Thesis-Draft2 - Copy Final

KOFORIDUA TECHNICAL UNIVERSITY
FACULTY OF APPLIED SCIENCE AND TECHNOLOGY
DEPARTMENT OF APPLIED MATHEMATICS
TIME SERIES ANALYSIS OF MALARIA IN THE NEW JUABEN MUNICIPALITY
BY
YAW BARIMAH
A PROJECT REPORT PRESENTED TO THE FACULTY OF APPLIED SCIENCE
AND TECHNOLOGY, DEPARTMENT OF APPLIED MATHEMATICS,
KOFORIDUA TECHNICAL UNIVERSITY IN PARTIAL FULFILLMENT OF THE
REQUIREMENT FOR THE AWARD OF BACHELOR OF TECHNOLOGY IN
STATISTICS.
JULY, 2018
STUDENT’S DECLARATION
I hereby declare that this submission is my own original work towards the award of Bachelor
of Technology in Statistics and that, to the best of my knowledge, It contains no material
previously published by another person nor material which has been accepted for the award
of any B-Tech in this University or elsewhere, except where due acknowledgement has been
made in the text.
Signature……. Date………………
YAW BARIMAH
i
SUPERVISOR’S CERTIFICATION
I do hereby certify that this project work was supervised by me in accordance with the rules
and regulations governing project work supervision in Koforidua Technical University. I
therefore have no reservations in recommending it for acceptance and use for the intended
purpose.
Signature: …………..……….. Date………………
APPIAH ANDANE AKABOHA
ii
DEDICATION
I cordially and loyally dedicate this piece of work to the Almighty God who by His Mercies
endowed me with knowledge and strength to undertake this academic journey successfully.
I again respectively dedicate it to my lovely families and loved ones.
iii
ACKNOWLEDGMENT
I wish to first and foremost express my sincere appreciation to the Almighty God for his
incalculable love, grace, protection and provision throughout the course of my studies in the
university.
I also wish to express my deepest appreciation to my supervisor Mr. Appiah Andane
Akaboha for supervising and making valuable contributions in the realization of the success
of this work.
Likewise, I thank all lecturers in the Faculty of Applied Science and Technology for the
immense academic knowledge impacted especially during the course work sessions.
Finally, I also express my deepest gratitude to my family, friends, loved ones and supervisor
again for supporting and assisting me in getting access to the needed data for this research
work.
iv
ABSTRACT
Malaria has become a financial burden on the Government of Ghana. In 2009, it cost the
country $760 million to treat malaria and the disease accounted for 10% of the country’s
GDP. There is evidence that North America and Europe have succeeded in eradicating the
disease completely because they implemented effective programmes. The case is different in
most developing countries which Ghana is not an exception. Despite the several control
programmes undertaking by government and private entities, little is known about the rate of
malaria infection in New Juaben municipality and Ghana as a whole. The study employed
quantitative approach which permitted the gathering of secondary data of the reported malaria
cases from 2010 to 2017 from the New Juaben Municipal health directorate for the analysis.
The data was imported into Minitab for analysis. The trend of monthly malaria cases within
the municipality as observed from fitted time series showed a linear decreasing trend.
ARIMA model selected was ARIMA (1, 1, 2). After the estimation of the parameters of
selected models, a series of diagnostic and forecasting accuracy tests were performed. The
general equation is written as 𝑌𝑡= 𝑌𝑡−1 + 𝜑(𝑌𝑡−1 − 𝑌𝑡−2 ) − ∅1 𝑒𝑡−1 − ∅2 𝑒𝑡−1 , where 𝑌𝑡 is the
maternal mortality, 𝜑 is the coefficient of the autoregressive with order 1, ∅1 is the
coefficient of moving average with order 1, ∅2 is the coefficient of moving average with
order 2 and 𝑌𝑡−1 is the fitted value for the model. Finally, the model was used to forecast
future malaria cases for the following 12 months period. The results of the forecasts revealed
a decreasing trend of malaria cases within the municipality. It was therefore recommended
that Government through the Ministry of Health should intensify its malaria reduction
strategies across the municipality to help sustain or further reduce the rate of malaria
infection.
v
TABLE OF CONTENTS
STUDENT’S DECLARATION
Error! Bookmark not defined.
SUPERVISOR’S CERTIFICATION
DEDICATION
ACKNOWLEDGMENT
ABSTRACT
TABLE OF CONTENT vi
LIST OF TABLES viii

LIST OF FIGURES iix
CHAPTER ONE
1.0 Introduction
1.1 Background of the study
1.2 Problem statement
1.3 General Objective
1.3.1 Specific Objectives
1.4 Research Questions
1.5 Significance of the study
1.6 Methodology
1.7 Organization of the Study
vi
CHAPTER TWO: LITERATURE REVIEW
2.1 The Nature of Malaria
2.2 Global Burden of Malaria Disease 8
2.3 Human Related Factors 9
2.4 Knowledge on Malaria
2.5 Socio-Economic Factors 11
2.6 Environmental Factors/Climate
2.7 Causes of Malaria
Error! Bookmark not defined.6
2.8 Life Cycle of Plasmodium
2.9 Pathogenesis of Malaria
2.10 Genetic Resistance 19
2.11 Clinical Features
2.11.1. Complications
2.12 Diagnosis of Malaria
2.13 Prevalence, Incidence and Determinants of Malaria
2.14 Management of Malaria
CHAPTER THREE :METHODOLOGY
3.1 Research Design
3.2 Sources of Data
3.3 Data Analysis
vii
3.4 Time Series Analysis
3.4.1 Lag
3.4.2 Differencing
3.4.3 Stationary and Non-Stationary Series
3.5 Components of Time Series 29
3.5.1 The Trend (D) 29
3.5.2 Seasonal Variation (S)
3.5.3 Cyclical Variations (C)
3.5.4 Irregular Variations (I)
3.6 Basic Assumptions in Time Series
3.6.1 Autocorrelation Function (ACF)
3.6.2 Partial Autocorrelation Function
CHAPTER FOUR : DATA ANALYSIS AND FINDINGS 33
4.1 Preliminary time series analysis
4.2 Trend of Malaria
4.3 Moving Average Plot of Malaria cases in New Juaben Municipality
4.4 Model checking (parameter Determination)
4.5 Model Estimation and Selection
4.6 Model Selection
4.7 Forecasting
viii
CHAPTER FIVE :DISCUSSIONS, CONCLUSIONS AND RECOMMENDATIONS 40
5.1 Findings
5.2 Conclusion
5.3 Recommendations
REFERENCES
APPENDIX
LIST OF TABLES
Table 4.1: The mode of AIC value

Table 4.2: Forecast of malaria cases for 2018
ix
LIST OF FIGURES
Figure 4.1: Trend analysis plot for Malaria Cases from January 2010 to December 2017
Figure 4.2: Moving average plot of malaria cases in New Juaben Municipality
Figure 4.3: Autocorrelation Function
Figure 4.4: Partial Autocorrelation Function
x
CHAPTER ONE
INTRODUCTION
1.1 Background of the study
Disease is an illness affecting humans, animals or plants, often caused by an infection. There
are different types of diseases affecting human beings which cause negative impact and
sometimes death on an individual. Malaria is a disease contracted in the society among the
children and the aged. It is recognized as a major and increasing threat to world health.
Malaria is a female-mosquito-borne infectious disease caused by a eukaryotic protist of the
genus plasmodium. Even though it affect all ages, Children under five years and pregnant
women are mostly vulnerable to this disease in Ghana and Africa as a whole. One million
people die of malaria each year, mostly in sub-Sahara Africa (Wiseman et al., 2000). The
disease is transmitted to humans through the bite of mosquitoes. The effects of malaria go
beyond mortality and morbidity as malaria endemic areas suffer dearly in terms of human
productivity and economic loss.
Malaria is a life-threatening disease causing havoc on the lives of the inhabitants in East and
West Africa. This was clearly shown in increased in morbidity, mortality and absenteeism in
the workforce as well as decrease in productivity. Concerning malaria, it has principally been
attributed to mosquito bites, lack of mosquito nets and insect repellents to the populace,
inadequate mosquito-control measures such as spraying insecticides inside houses and
draining standing water where mosquitoes lay their eggs.
According to AngloGold (2004), the company realized that an estimated average of 11,000
malaria cases per month was recorded according to the municipal health authority. An
additional 6,800 cases were reported by the nine medical service. At any point in time, 20%
1
of the workforce had malaria and the average time off work due to malaria ailment was
between two and three days. Extending the trend to the whole country clearly indicated that
the disease had gained grounds and therefore called for actions to be taken to reverse the
increasing trend of the disease.
Again, in 2008, there were 247 million cases of malaria and nearly one million deaths –
mostly among children living in Africa. In Africa a child dies every 45 seconds of Malaria,
the disease accounts for 20% of all childhood deaths.
It must be emphasized that malaria has been one of the most prominent and ancient diseases
which has been profiled and studied. It has been one of the greatest burdens to mankind, with
a mortality rate that is unmatched by any other modern disease other than tuberculosis
(Sudhakar et at., 2007).
It remains the leading cause of death in children under five years in Africa (Houeto et al.,
2007).
Malaria is one of the leading killer diseases in the tropical and subtropical countries. It
therefore poses a serious health problem to these countries including Ghana. This disease is
frequently called disease of the poor because its prevalent rate is very high in the poorest
continent and in the poorest countries (Worral et al., 2003).
According to Houeto et al., (2007) successes have been made in the areas of prevention and
treatment through the adoption of artemisinin combined therapy (ACT). The use of
insecticide treated nets has helped to reduce morbidity and mortality rates. But the disease
still poses a threat to our part of the world. According to Nchinda, (2005) sub-Sahara Africa
was never part of the global malaria eradication programme because the period coincided
with colonial and immediate Postcolonial period and so the indigenous had little or no power
2
to initiate and sustain an eradication programme. Most studies have concentrated on the
factors affecting the prevalent rate of this dreadful disease and some of the control measures
which have already been initiated to control the disease. There is combination of factors
which contribute to the resurgence of this disease in Ghana and Africa as a whole. There is
therefore the need for finding a statistical model for forecasting the reported cases as well as
deaths as a result of malaria in the municipality. The developed model will help the Ghana
Health Service to allocate adequate resources to efficiently control the disease.
1.2 Problem statement
Malaria has become a financial burden on the Ghana health service in Ghana. During the
launched of the Nationwide Malaria Control Programme by Zoomlion Ghana Limited at
Koforidua in 2009, the minister of health, Dr Sipa Yankey, said that it cost the country $760
million to treat malaria and the disease accounted for 10% of the country’s GDP. There is
evidence that North America and Europe have succeeded in eradicating the disease
completely because they implemented effective programmes. The case is different in most
developing countries which Ghana is not an exception. Despite the several control
programmes undertaking by government and private entities, there is still high prevalence
rate of malaria in New Juaben municipality and Ghana as a whole.
Some studies have used mathematical tools to find out whether there is significant impact of
the malaria intervention programme on the reported malaria cases over the years. The
average monthly reported cases of the disease before and after the intervention would be
compared to find the effectiveness of the intervention. However, there has been paucity of
studies to understudy the long-term behaviour of the disease in other to device suitable
strategies to combat the disease at the municipality level. There is therefore the need for
suitable time series model to be formulated that would help in finding the general trend of the
3
malaria reported cases as well as deaths that arise from malaria disease. Modelling the
reported cases through correlated models, such as time series would allow greater part of data
behavioural dynamics to be adjusted into a single equation and future malaria reported cases
and deaths would be estimated based on this. This would benefit the municipality in areas of
planning, reserving resources and performing more efficient and timely control of the disease.
1.3 General Objective
The general objective of the study is to statistically analyse malaria cases in the New Juaben
municipality of eastern region of the Republic of Ghana.
1.3.1 Specific Objectives
In order to achieve the objective of the study the research will work along these specific
objectives:
i. To develop appropriate time series model for malaria cases in the New Juaben
Municipality
ii. Determine the trend of malaria cases in the New Juaben Municipality from 2010-
2017
iii. To forecast malaria cases in the Municipality for the next 5 years period.
1.4 Research Questions
The research issues that informed the researchers include the following:
i. How is the underlying behaviour of malaria cases in the New Juaben Municipality?
ii. What is the appropriate time series model for malaria cases in the New Juaben
Municipality?
4
iii. How will the malaria situations in New Juaben Municipality be in the next 5 years
period?
1.5 Significance of the study
Several programmes have been initiated in this country to combat this disease of the poor but
its prevalent rate is still high and accounted for 40% out of the 70% communicable diseases
in Ghana in 2008. The National Malaria Control Programme (NMCP) is the mother agent for
controlling the disease in this country. Other companies like Zoomlion Company Limited
have also joined in the fight against the disease.
The ability to predict would provide a mechanism for the NMCP authority, non-
governmental organizations and other health-care services to respond to outbreak in timely
fashion and minimizing its effect. At the end of the research, the programme would have to
be repackaged or recommended for adoption by the National Malaria Control Programme and
other non-governmental organizations to help control the disease.
The study will provide policy makers with the evidence and the idea of possible future values
of malaria, and thus help them to revise their malaria death threat intervention strategies so as
to maintain and sustain the rates, or to reduce further, if the 2017 rate is found to be far from
the target value. All these will ensure that Ghanaian are healthier and grow to realize their
talents and potentials and thus help maintain a strong labour force for the future that will
continue with the developmental program of the nation Ghana as well as reducing the
national expenditure budgets on health care.
Finally, this study will add to exiting literature regarding malaria death intervention
strategies. This study will further serve as reference for elaborative academic research, adding
to literature already in existence.
5
1.6 Methodology
The study will employ quantitative approach which will permit the gathering of secondary
data of the reported malaria cases from 2010 to 2017 from the New Juaben Municipal health
directorate for the analysis. The data will be imported into Minitab for analysis. Descriptive
analysis of the data will be done to understand the malaria situations in the Municipality.
Further analysis of the data will be performed using time series analysis in order to
decompose the data into explainable components which will allow forecast to be made for
future cases.
1.7 Organization of the Study
This study will comprise of five Chapters. The first chapter will focus on background of the
research, problem statement, research objective and questions. It will also discuss the
significance of the study and a brief discussion on the research methodology that will be
adopted for the study and finally ends with the organization of the report.
Second chapter basically reviews literature on key concepts and terminologies relevant to the
research issue.
Third chapter will deal with the method of the research: profile, design, source of data and
method of data analysis technique.
Fourth chapter will also deal with the analysis of data collected and presentation of findings
as well as interpretation of the results.
The final chapter will engage on discussions, conclusions and recommendations.
6
CHAPTER TWO
LITERATURE REVIEW
This chapter deals with the review of existing literature related to the topic under study. It
covers the scope of malaria cases associated with the diseases. The typical examples of this
disease is what is happening in countries which have experienced the outbreak of the disease
and the mechanisms adapted to help prevent the disease and its related effects on gender.
2.1 The Nature of Malaria
From the study of Nayyar, et al (2006), Malaria is a mosquito-borne infectious disease of
humans and other animals caused by parasitic protozoans(a type of unicellular
microorganism) of the genus Plasmodium. Commonly, the disease is transmitted via a bite
from an infected female Anopheles mosquito, which introduces the organisms from its saliva
into a person's circulatory system. In the blood, the protists travel to the liver to mature and
reproduce. Malaria causes symptoms that typically include fever and headache, which in
severe cases can progress to coma or death. The disease is widespread in tropical and
subtropical regions in a broad band around the equator, including much of Sub-Saharan
Africa,Asia, and the Americas. Five species of Plasmodium can infect and be transmitted by
humans. The vast majority of deaths are caused by P.falciparum and P.vivax, while P.ovale,
and P.malariae cause a generally milder form of malaria that is rarely fatal. The zoonotic
species P.knowlesi, prevalent in Southeast Asia, causes malaria in macaques but can also
cause severe infections in humans. Malaria is prevalent in tropical and subtropical regions
because rainfall, warm temperatures, and stagnant waters provide habitats ideal for mosquito
larvae. Disease transmission can be reduced by preventing mosquito bites by using mosquito
7
nets and insect repellents, or with mosquito-control measures such as spraying insecticides
and draining standing water.
Malaria is typically diagnosed by the microscopic examination of blood using blood films,
or with antigen-based rapid diagnostic tests. Modern techniques that use the polymerase
chain reaction to detect the parasite's DNA have also been developed, but these are not
widely used in malaria-endemic areas due to their cost and complexity. The World Health
Organization has estimated that in 2010, there were 219 million documented cases of
malaria. That year, the disease killed between 660,000 and 1.2 million people, many of
whom were children in Africa. The actual number of deaths is not known with certainty, as
accurate data is unavailable in many rural areas, and many cases are undocumented. Malaria
is commonly associated with poverty and may also be a major hindrance to economic
development. Despite a need, no effective vaccine exists, although efforts to develop one are
ongoing. Several medications are available to prevent malaria in travellers to malaria-
endemic countries (prophylaxis). A variety of antimalarial medications are available. Severe
malaria is treated with intravenous or intramuscular quinineor, since the mid-2000s, the
artemisinin derivative artesunate, which is superior to quinine in both children and adults
and is given in combination with a second anti-malarial such as mefloquine. Resistance has
developed to several antimalarial drugs; for example, chloroquine-resistant P. falciparum
has spread to most malarial areas, and emerging resistance to artemisinin has become a
problem in some parts of Southeast Asia (Nayyar, et al, 2006).
2.2. Global Burden of Malaria Disease
Malaria is considered the most consequential parasitic infection in humans. There are as
many as 350-500 million clinical episodes per year worldwide (UNICEF, 2003) and while
most estimates of mortality caused by malaria lie at around 1 million deaths per year (Snow
8
et al. 2005), some calculations go as high as 3 million (Breman et al, 2004). Almost all of
these deaths occur in children (Phillips, 2001), living in countries in sub-Saharan Africa
(SSA) (Ukoli, 1990) where 25% of all childhood mortality below the age of five (about
800,000 young children (Shepard et al., 1991) is attributable to malaria. Of those children
who survive cerebral malaria, more than 15% suffer neurological deficits (NIMR 2006, &
Schönfeld et al, 2007) which include weakness, spasticity, blindness, speech problems and
epilepsy. Where such children are poorly managed and do not have access to specialized
educational facilities, these deficits may interfere with future learning and development
Children under the age of five years are at highest risk for malaria because they have not yet
acquired protective immunity. People with semi-immunity are infected, but do not get a
severe disease as a rule. In stable transmission areas new-borns are protected by the IgM
antibodies of their mother and through breastfeeding. After three months children have a
higher susceptibility for an infection with the parasite. In high transmission areas this time
period lasts until the age of 3-5 years. In areas with a seasonal transmission the period can
last 10 years. Without re-infection the acquired immunity can disappear in a matter of years
(Eddleston et al. 2008).Furthermore children under five years of age experience the biggest
malaria burden because they are often super-infected with other parasites and/or that they
often suffer from nutritional deficiencies. These lead to a weaker immune system, which
leads to a higher susceptibility for malaria. Moreover, a malaria infection and malnutrition
are reasons for an increasing anaemia burden in children (Greenwood et al.1991).
2.3 Human Related Factors
There is a large amount of data on malaria related morbidity and mortality in children under
five. (Menard et al, 2010) suggested that, the risk of infection and its severity is lower in the
first few months of life. Reasons for this are complex but probably include transmission of
9
protective antibodies across the placenta, the presence of red cells containing Hbf – which
are relatively resistant to malaria infection, breast feeding and lack of exposure (Hviid,
&Staalsoe T.2004).The protective effect of maternal antibody is likely to be less when
effective malaria control is achieved and the overall level of malaria infection declines. In
lower transmission settings clinical malaria is spread more widely across the age groups. In
such settings, occupational issues may become more important than age; this is especially
true where mosquitoes which transmit malaria bite outdoors away from dwellings. Forest
workers in south-east Asia are one example of this phenomenon (Erhartet al, 2004). In these
settings young adults, especially males, may be more at risk than children; because they are
the group at most risk from being bitten by forest dwelling vectors (Dysoley et al, 2008).
Furthermore, this information was supported by a study that was done in Kenya that
explored factors affecting use of permethrin-treated bed nets during a randomized controlled
trial found that children less than five years of age were less likely to use nets compared to
the individual adults (Alaii et al, 2003).
2.4 Knowledge on Malaria
A number of studies have investigated differences in knowledge and reported health seeking
behaviour between men and women. Most found either no difference or those women had
more limited decision-making and financial power to act. This was associated with failures
and delays in seeking treatment, with differential understanding of malaria between men and
women, and differential health-seeking behaviour. Women delayed seeking care until men
were available, while men were less willing to spend on child health. (Al-Taiar et al 2007
&Oberlander and Elverdan 2000). These differences are critical when considering the main
child-caring role of women and children‘s increased vulnerability to malaria. Furthermore,
according to (Minja et al., 2001), it was stated that knowledge, attitude and behaviour
10
practices regarding malaria were shown to influence the ITN ownership. Some other studies
that were done by
(Victora, 2003, Nganda 2004&Magesa et al., 2005) reported that knowledge of the problem;
affordability and accessibility are among major obstacles for the ITN ownership and use.
Net ownership has also been related to the educational levels of household members. This is
a complicated relationship since educational attainment can have the impact on an
individual‘s ability to understand and access information regarding malaria prevention
methods. In Malawi it was found that net ownership was less common in households where
the head/caretaker had not completed primary school and in homes where the house had
mud walls or a grass roof (Holtz et al., 2002). This is being supported by (Nuwaha, 2001)
with evidence that educational attainment is associated with malariaspecific knowledge and
uptake of preventive measures. Rhoida Y et al 2004 concluded that the success in
implementing preventive interventions amongst pregnant women in Tanzania is thus likely
to be determined in part by awareness of malaria and the strategies available to prevent it. In
order for the ITN distribution programme to succeed, the knowledge gaps, practices and
attitudes that may negatively influence the intervention uptake.
2.5 Socio-Economic Factors
Socioeconomic conditions of the community have direct bearing on the problem of malaria.
Ignorance and impoverished conditions of people contribute in creating source and spread of
malaria and hinder disease control strategy (Collins et al, 1997&Yadav et al., 1999). This
was also evidenced by Filmer 2002, that high costs of malaria treatment may lead to delays
in treatment seeking behaviour, whereby he found that the poorest groups in a society did
not seek care as much as the non-poor, and did so at lower level public facilities. Economic
inequities in areas such as the control of household resources also affect access to ITNs. In
11
one study in Benin, many women explained that since they were financially dependent on
their husbands, they were unable to purchase an ITN for themselves and their children
unless their husbands prioritized the use of bed nets (Krause G, et al 2000)).The study also
revealed that when women did earn an income and had control over this income, they were
much more likely than men to purchase an ITN for their household.(Krause G, et al 2000)
Alnwick 2000,also observes that malaria afflicts primarily the poor, who tend to live in
dwellings that offer little or no protection against mosquitoes. Furthermore,
(Brown, 1997) notes that the continuation of brutal poverty and hunger in much of the world
is undoubtedly linked to large numbers of unnecessary deaths from malaria. A survey in
Zambia also found a substantially higher prevalence of malaria infection among the poorest
population groups (Roll Back Malaria, 2001). According to (Makundi et al,2007) it was
reported that the burden of malaria is greatest among poor people, imposing significant
direct and indirect costs on individuals and households and pushing households into in a
vicious circle of disease and poverty. Furthermore, vulnerable households with little coping
and adaptive capacities are particularly affected by malaria. Households can be forced to sell
their food crops in order to cover the cost of treatment (Wandiga et al, 2006.) Depleting
household resources and leading to increased food shortages, debts, and poverty for the
poorest households. The costs of malaria are highly regressive, with the poorer households
spending a significantly higher proportion of their income on the on the treatment of malaria
than their least poor counterparts.
In Ghana, both direct and indirect costs associated with a malaria episode represent a
substantial burden on poorer households. A study found that while the cost of malaria care
was just 1 per cent of the income of the rich, it was 34 percent of the income of poor
households (Akazili, 2002). Similarly, (Kuate,1997) found that the burden of illness rests
disproportionately on the economically disadvantaged women who were not employed,
12
women living in poor neighbourhoods, and those living in households without modern
facilities. This is being supported by a study that was done on the use of ITNs whereby it
was only recently appreciated that a net treated with insecticide offers much greater
protection against malaria. Unfortunately, the commercial price of nets and insecticide is
beyond the poorest income groups of the population (World Health Organization, 2003).
Ziba et al 1994: found that in Malawi, use of malaria prevention measures (bed nets,
insecticides, mosquito coils, other insect repellents, burning leaves, etc.) was income
dependent. In households where the head earned a larger than average income, use of
commercial methods (mosquito coils, insecticide spray, bed nets) was more common. Use of
inexpensive and less effective, natural methods (burning leaves, dung, or wood) was
associated with lower income. Occupational and cultural differences related to undertaking
activities likely to lead to malaria transmission; and when malaria is acquired, access to
health services is more mixed and varies considerable across different cultural settings.
Alternatively, if a household only has one bed net, priority may be given to the male head of
the household as he is often considered the primary breadwinner (Krause, et al 2000). Before
the ITNs project started in Bagamoyo, Tanzania, it was reported that it was mainly the adult
men who used the nets, followed by women and children under two who sleep with their
mothers, while elder children were frequently the last to gain access (Makemba et.al.1995).
ITNs availability and efficacy. The use of insecticide-treated nets is currently considered one
of the most cost-effective methods of malaria prevention in highly endemic areas Tanzania
being included. This has been achieved through free distribution of long lasting insecticidal
nets (ITNs) that has been conducted through campaigns, public health facilities, faith-based
organisations (FBO), and non-governmental organisations (NGOs) with the goal of
achieving universal access for the at-risk population of children under age five and pregnant
women. The use of ITNs in Tanzania has increased markedly over the past few years.
13
Statistics indicate that the proportion of households with at least one untreated net has
increased from 14% in 2001 to 58% in 2005.ITNs coverage is estimated to have reached
63% of households with at least one ITN and 25%b and 26% of children under 5and
pregnant were respectively sleeping under an ITN. However, more effort is needed to further
increase this to reach the globally agreed target of 80(2007-2008
YHMIS). According to the (MoHSW 2006) report, Insecticide-treated mosquito nets (ITNs)
used for protection against mosquito bites has proven to be a practical, highly effective, and
cost-effective intervention against malaria. This was also supported by (Roll Back Malaria,
2005) which reported that ITNs reduces human contact with infected mosquitoes and have
been shown to be an effective malaria prevention measure. In addition to the direct benefit to
the individual, ITNs use offers a protective benefit for the entire community (Teklehaimonot
et al., 2007). Studies examining ITN‘s efficacy suggest a significant reduction in malaria
episodes. If used universally, ITNs could prevent approximately 7% of the global under-five
mortality (Jones et al, 2003). The UNICEF corroborates that under-five mortality rates could
be reduced by about 25-30% if all young children in malaria endemic areas were protected
by treated bed nets at night. This was also evidenced by (Lengeler‘s,2004) review which
demonstrates the efficacy of ITNs in both stable and unstable transmission areas where by it
was documented the wide spread use of ITNs resulted in an overall reduction in mortality of
19 percent, protected against anaemia, and had a substantial impact on mild disease
episodes. Another One large-scale rural study in Tanzania found that ITNs and untreated
nets reduced mortality of children one month to four years, with protective efficacies of 27
and 19 percent, respectively (Armstrong and Schellenberg,2002). Hill et al wrote in 2006
that despite of all these efforts and its efficacy, only 3% of African children sleep under
these treated nets while only about 20% sleep under any other kind of nets. This however
may account for the high rate of mortality amongst children due to malaria and its related
14
problems. Despite the evidence that the use of ITNs decreases malaria-related morbidity and
mortality, the use of ITNs in sub-Saharan Africa remains relatively low.
2.6 Environmental Factors/Climate
Malaria is governed by a large number of environmental factors, which affect its
distribution, seasonality and transmission intensity (Snow et al, 1999). Climate and
environmental conditions greatly affect the transmission and incidence of malaria, by
influencing primarily the abundance and survival of vectors and parasites, and also exposure
of humans and other hosts. (Lafferty, 2009). The most important environmental factors for
malaria transmission have to do with conditions for Anopheles mosquito breeding and
survival water in which they can breed, and minimum temperatures and humidity to allow
them to survive long enough for the vector stage of the parasite‘s life cycle to be completed
usually about ten days. These factors are influenced by climate, as well as by topography
and soil conditions, drainage, vegetation cover, land use and water all of which vary greatly
depending on local conditions. As such, changes in climate and land use such as water
management, agriculture, urbanization, and deforestation can lead to significant increases or
decreases in malaria transmission, depending on local contexts.
(Reiter P.2001) Some agricultural practices facilitate the spread of vector-borne diseases.
Also, the presence of cattle in marshy areas results in the creation of hoof prints that
potentially offer ideal conditions for mosquito breeding. Within manmade malaria,
excluding the migration of non-immunes to endemic areas, the most important impacts on
transmission are probably brought about by water resource development and land use
change. Human modification to the environment also can create larval development sites
and malaria (Denise et al, 2003). This may especially be true for manmade malaria in
15
which man by his farming activity or any other activity may create the environments which
suit mosquito breeding and protective measures may be widely distributed. The
identification of predictors of malaria incidence could provide a useful means of identifying
targets for intervention of manmade malaria (Ghebreyesus et al., 2000).
Furthermore Utilization of ITNs has, however, been found to vary with Binka et al showed
seasons of the year and acceptability of the nets in terms of size, colour and shape. That the
time of the year during which the nets are delivered affects use. 99% of the net recipients
were found to use the nets during rainy season, while only 20% used it during the dry season
this was evidenced by a study which was done in Burkina Faso of which reported a decreased
use of bed nets during the dry season due to a perceived lower risk of mosquito bites and the
practice of sleeping outdoors (Frey et al, 2006). Malaria control strategies need to consider
how changing environmental conditions may be linked to an increase or decrease of malaria
transmission. Opportunities exist for integrating environmental management interventions
into vector control strategies in order to reduce malaria risk.
2.7. Causes of Malaria
Malaria parasites belong to the genus Plasmodium(phylum Apicomplexa). In humans,
malaria is caused by P.falciparum,P.malariae,P.ovale,P.vivaxand P.knowlesi. Among those
infected, P. falciparum is the most common species identified (~75%) followed by P. vivax
(~20%). Although P. falciparum traditionally accounts for the majority of deaths, recent
evidence suggests that P. vivax malaria is associated with potentially life-threatening
conditions about as often as with a diagnosis of P. falciparum infection. P. vivax
proportionally is more common outside of Africa (Okenul, 2003). There have been
documented human infections with several species of Plasmodium from higher apes;
16
however, with the exception of P. knowlesi—a zoonoticspecies that causes malaria in
macaques—these are mostly of limited public health importance (Okenul, 2003).
2.8 Life Cycle of Plasmodium
All types of malaria have a similar life cycle. Sporozoites, the infectious form of the malaria
parasite, are injected into a human host through the saliva of an Anopheles mosquito. These
sporozoites enter the liver cells within minutes, take on a new form, and multiply. When the
liver cells rupture, blood stage parasites—known as merozoites—are released. Each
merozoite invades a red blood cell, and for two days multiplies into more merozoites. The
red blood cell full of merozoites ruptures to release more merozoites. It is this stage of the
life cycle that causes disease and, too often, death. Some merozoites change into the form
called gametocytes, which do not cause disease but remain in the blood until they are cleared
by drugs or the immune system, or taken up by the bite of a mosquito. In the mosquito's
stomach a "male" gametocyte fertilizes a "female" to form an egg, or oocyst, which matures
into thousands of sporozoites that swim to the mosquito's salivary glands to be injected into
another human at the next bite (Gallup, 2001).
2.9 Pathogenesis of Malaria
In humans, malaria is caused by four species of the plasmodium protozoa (single celled
parasites) – plasmodium falciparum, plasmodium vivax, plasmodium ovale and plasmodium
malariae. Of these species plasmodium falciparum accounts for the majority of infections
and is the most lethal. Several studies have been done on different aspects of the disease,
from parasitology to finding a cure with drugs (chemotherapy) and to eradication of the
disease by the use of insecticide treated net and insecticides. Rashed4 conducted a study
which was aimed at determining the effect of Permethrin insecticide treated nets (PITN) use
17
on the incidence of febrile episodes and non-household malaria expenses in Benin. The
study found out that, the use of PITNs decreased the risk of developing malaria by 34% in
children in the rural areas; meanwhile, PITN use did not reduce prevention and treatment
expenses. In a parasitology laboratory, malaria was found to be the major killer of paediatric
illness and death in Kinshasa (Coene 1991).
In view of this, the treatment of fevers as malaria with chloroquine is no longer acceptable
because the plasmodium falciparum had a resistance to chloroquine. According to the study,
the differences in endemicity of malaria that existed between the various parts of town had
to be taken into consideration alongside the ecological and socio-economic factors that
underlie when planning for estimation of potential control methods. The behavioural risk for
malaria in the Machodinho resettlement area in the Amazonian forests of Brazil was
examined (Castilla and Sawyer 1993). Analysis of the study suggested that economic status
and knowledge of the importance and behaviour of the mosquito in transmitting malaria are
significant factors in determining prevalence risk irrespective of whether preventive
precautions, for example, dichlorodiphenyl trichloroethane (DDT) spraying of houses and
cleaning of vector breeding sites are to be undertaken in the endemic areas. However, the
researchers found out that a higher economic status combined with better knowledge of the
vector and DDT spraying of houses decreased the risk of infection. They suggested that a
more positive implication is that control programmes must work harder and more intensively
on behalf of poorer people especially migrants in order to diminish the disease burden for
them. Sharma and colleagues (2001) carried out a study on the socioeconomic factors as
well as on the human behaviour towards malaria on cross section of the Sundargarh district
in India. They argued that poor socioeconomic status and socio-cultural factors play an
important role in maintaining high degree of malaria transmission. They found that human
behaviours such as location of hamlets, type of malaria transmitted, sleeping habits, and
18
outdoor activities after dusk, poor knowledge about the disease and treatment seeking
behaviour are of great significance as determinants of malaria transmission.
Malaria is also a major problem in Papua New Guinea as it accounts for a high proportion of
sickness and death. This is because in addition to human suffering, it also put severe stress
on the health facilities and directly hinders economic growth. It has been suggested that a
malaria vaccine would be best, most cost effective and safe public health measure to reduce
the burden of malaria (Reeder 2001). Whitty and Allan (2004) contend that the serious threat
posed by the spread of drug-resistant malaria in Africa has been widely acknowledged.
Chloroquine resistant malaria is now almost universal and resistant to successor drug,
sulfadoxine-pyrimethamine (SP) is growing rapidly. If the question of cost of treatment is
not successfully addressed this could lead to adverse result from the deployment of
combination therapy as a first-line treatment. Adverse effect of costly treatment ranges from
increase in delays in infected individuals presenting themselves to the health care facilities
for treatment to exclusion of the poorest malaria sufferers from receiving treatment
altogether.
2.10. Genetic Resistance
According to a 2005 review, due to the high levels of mortality and morbidity caused by
malaria—especially the P. falciparum species—it has placed the greatest selective pressure
on the human genome in recent history. Several genetic factors provide some resistance to it
including sickle cell trait, thalassaemiatraits, glucose-6-phosphate dehydrogenase deficiency,
and the absence of Duffy antigens on red blood cells (Kwiatkowski, 2005). The impact of
sickle cell trait on malaria immunity illustrates some of the evolutionary trade-offs that have
occurred because of endemic malaria. Sickle cell trait causes a defect in the haemoglobin
molecule in the blood. Instead of retaining the biconcave shape of a normal red blood cell,
19
the modified haemoglobin molecule causes the cell to sickle or distort into a curved shape.
Due to the sickle shape, the molecule is not as effective in taking or releasing oxygen.
Infection causes red cells to sickle more, and so they are removed from circulation sooner.
This reduces the frequency with which malaria parasites complete their life cycle in the cell.
Individuals who are homozygous(with two copies of the abnormal haemoglobin beta allele)
have sickle-cell anaemia, while those who are heterozygous (with one abnormal allele and
one normal allele) experience resistance to malaria. Although the shorter life expectancy for
those with the homozygous condition would not sustain the trait's survival, the trait is
preserved because of the benefits provided by the heterozygous form (Kwiatkowski, 2005).
2.11 Clinical Features
The signs and symptoms of malaria typically begin 8–25 days following infection; however,
symptoms may occur later in those who have taken antimalarial medications as prevention.
Initial manifestations of the disease—common to all malaria species—are similar to flu-like
symptoms, and can resemble other conditions such as septicaemia, gastroenteritis, and viral
diseases. The presentation may include headache, fever, shivering, joint pain, vomiting,
haemolytic anaemia, jaundice, haemoglobin in the urine, retinal damage, and
convulsions(Sherman, 1998).
The classic symptom of malaria is paroxysm—a cyclical occurrence of sudden coldness
followed by shivering and then fever and sweating, occurring every two days (tertian fever)
in P. vivax and P. ovale infections, and every three days (quartan fever) for P. malariae. P.
falciparum infection can cause recurrent fever every 36–48 hours or a less pronounced and
almost continuous fever. Severe malaria is usually caused by P. falciparum (often referred to
as falciparum malaria). Symptoms of falciparum malaria arise 9–30 days after infection.
Individuals with cerebral malaria frequently exhibit neurological symptoms, including
20
abnormal posturing, nystagmus, conjugate gaze palsy(failure of the eyes to turn together in
the same direction), opisthotonus, seizures, or coma(Sachs, 2001).
2.11.1. Complications
There are several serious complications of malaria. Among these is the development of
respiratory distress, which occurs in up to 25% of adults and 40% of children with severe P.
falciparum malaria. Possible causes include respiratory compensation of metabolic acidosis,
noncardiogenic pulmonary oedema, concomitant pneumonia, and severe anaemia. Although
rare in young children with severe malaria, acute respiratory distress syndrome occurs in 5–
25% of adults and up to 29% of pregnant women. Coinfection of HIV with malaria increases
mortality. Renal failure is a feature of black water fever, where haemoglobin from lysedred
blood cells leaks into the urine (WHO, 1999).
Infection with P. falciparum may result in cerebral malaria, a form of severe malaria that
involves encephalopathy. It is associated with retinal whitening, which may be a useful
clinical sign in distinguishing malaria from other causes of fever. Splenomegaly, severe
headache, hepatomegaly (enlarged liver), hypoglycemia, and hemoglobinuria with renal
failure may occur. Malaria in pregnant women is an important cause of stillbirths, infant
mortality and low birth weight, particularly in P. falciparum infection, but also with P. vivax
(WHO, 2005).
2.12 Diagnosis of Malaria
Owing to the non-specific nature of the presentation of symptoms, diagnosis of malaria in
non-endemic areas requires a high degree of suspicion, which might be elicited by any of the
following: recent travel history, enlarged spleen, fever, low number of platelets in the blood,
and higher-than-normal levels of bilirubin in the blood combined with a normal level of
21
white blood cells. Malaria is usually confirmed by the microscopic examination of blood
films or by antigen-based rapid diagnostic tests(RDT) (Perking, 2007). Microscopy is the
most commonly used method to detect the malarial parasite; about 165 million blood films
were examined for malaria in 2010. Despite its widespread usage, diagnosis by microscopy
suffers from two main drawbacks: many settings (especially rural) are not equipped to
perform the test, and the accuracy of the results depends on both the skill of the person
examining the blood film and the levels of the parasite in the blood. The sensitivity of blood
films ranges from 75–90% in optimum conditions, to as low as 50%.
Commercially available RDTs are often more accurate than blood films at predicting the
presence of malaria parasites, but they are widely variable in diagnostic sensitivity and
specificity depending on manufacturer, and are unable to tell how many parasites are present
(WHO, 2004).
In regions where laboratory tests are readily available, malaria should be suspected, and
tested for, in any unwell patient who has been in an area where malaria is endemic. In areas
that cannot afford laboratory diagnostic tests, it has become routine to use only a history of
subjective fever as the indication to treat for malaria; a presumptive approach exemplified
by the common teaching "fever equals malaria unless proven otherwise". A drawback of this
practice is over diagnosis of malaria and mismanagement of non-malarial fever, which
wastes limited resources, erodes confidence in the health care system, and contributes to
drug resistance. Although polymerase chain reaction-based tests have been developed, these
are not widely implemented in malaria-endemic regions as of 2012, due to their complexity
(Gallup et al, 2001).
22
2.13 Prevalence, Incidence and Determinants of Malaria
It should be noted that epidemic malaria is derived from interactions of vectors, parasites
and various environmental and anthropogenic determinants. Malaria epidemic afflict
immunological vulnerable populations, straining the capacity of health facilities and causing
case fatality rates to increase five-fold or more during outbreaks. The demographic profile
may translate into larger economic consequences, although the full economic impact of
epidemic malaria remains undefined. A study was conducted in Benin on how to conceive
and establish the importance of economic factors that contributed to malaria transmission
(Mensah and Kumaranayake 2004).
According to the study, despite the endemic malaria situations, there was still little
understanding of the relative importance of economic factors that contribute to people
acquiring the disease in communities where malaria was endemic. The researchers
contended that, predisposing characteristics of household’s heads such as age, knowledge of
malaria, education and size of household significantly affect the incidence of malaria as
anticipated by economic theory. A study by Asenso – Okyere (1994) on malaria in 4 districts
namely Kojo Ashong, Barekese, Barekuma and Oyereko all from the Greater Accra Region
of Ghana revealed that factors that were perceived as causing malaria are malnutrition,
mosquitoes, excessive heat, excessive drinking, flies, fatigue, dirty surroundings, unsafe
water, bad air and poor hygiene. Almost all the adolescents at that time had no idea how the
disease was spread from person to person, while the symptoms of clinical malaria was also
frequently considered to be yellowish eyeball, chills and shivering, headache, a bitter taste,
body weakness and yellowish urine, the study added.
23
2.14 Management of Malaria
Yeboah-Antwi and colleagues (1997) examined the extent to which district health teams in
Kintampo in the Brong Ahafo Region of Ghana could reduce the burden of malaria, which is
a major cause of mortality and morbidity in a situation where severe resource constraints
existed. It was found out that, compliance improve by approximately 20% in both adults and
children but there was improvement to care about 50% for example in cost to patients,
waiting time at dispensaries and drug wastage at facilities. Another case study in Ghana
sought to compare household’s data on acute morbidity and treatment seeking behaviour in
two districts with the use of health facility data (Agyepong and Kangeya Kayonda 2004).
For every case of febrile illness seen in the health facilities there were approximately 4-5
cases in the community, hence they concluded that every febrile episodes especially in
children be treated with an anti-malarial drug. Since several countries extend malaria
treatment to include the community and the home through public and private, formal and
informal sectors, the need for more comprehensive estimates becomes urgent. Appawu and
colleagues (2004) studied malaria transmission dynamics in the KassenaNankana District, a
site in northern Ghana proposed for testing malaria vaccines. Intensive mosquitoes sampling
was done for one year using human landing catches in three micro-ecological sites that is
irrigated, lowland and rocky highlands.
Transmission was highly seasonal and the heaviest transmission occurred from June to
October. The intensity of transmission was higher for people in the irrigated communities
than the non-irrigated ones. Approximately 60% of malaria transmission in KND occurred
indoors during the second half of the night, peaking at daybreak between 04.00 to 06.00
hours.
24
CHAPTER THREE
METHODOLOGY
This chapter identifies the various methods adopted in collecting data for analysis and it
covers, research design, sources of data and method of data presentation. Additionally, the
chapter reviews the time series model used in the further analysis.
3.1 Research Design
Research design outlines research study, which indicates what the researcher will do from
writing the research question and its operational implications to the final analysis of data.
According to Trochim (2006), research design can be thought of as the structure of the
research; it is the “glue” that holds all of the elements in a research project together. It is a
plan of what to gather, from whom, how and when to collect the data, and how to analyse the
data obtained. Research design constitutes decision regarding what, why, where, when and
how concerning an inquiry or a research study (Sekaran, 2011). Bryman (2001) points out
that when designs of the two approaches namely quantitative and qualitative are combined,
they help in supporting and collaborating each other.
This study will use quantitative approach. The quantitative research tries to find answers to
questions through analysis of quantitative data, i.e., the data shown in figures and numbers.
Quantitative research clearly and precisely specifies the independent, mediating and the
dependent variables under investigation. Quantitative methods also have the ability to use
smaller groups of people to make inferences about larger groups that would be prohibitively
expensive to study (Holton & Burnett, 1997; Matveev, 2002). Based on the nature of the
research objectives, a quantitative research approach is most suitable for this study.
25
3.2 Sources of Data
There are two basic sources of data: thus, data gathered directly from field and use for an
intended purpose, it is usually referred as primary data or first-hand information. However
then if a researcher uses information which already existed for other purposes rather than the
current usage, it could be referred as secondary data. It must be noted that this study used
data from both sources.
In an investigation of this nature the researcher needed to first review literature or already
existing knowledge on the subject matter. Hence the researcher gathered information from
articles, books, reports and internet sources to enable the understanding of theoretical
concepts and definitions.
However, after building the theoretical concepts there was the need to practically examine
assertions and claims. This was then done through the gathering of data from Ghana Health
Service District Directorate.
3.3 Data Analysis
The data collected will be quantitative in nature. The data handling exercise will start entry
to Minitab. The computer software Minitab will be used to analyse the data because is among
the most widely used statistical software in the social sciences to analyse trend. This package
provides users with substantial increase in the ease and flexibility with which they can
approach their day to day use of the computer to generate results for easy interpretation.
Further, this will make the data more logical and ensure easy understanding of the analyses.
The findings will be scrutinised for consistency with the research questions and then
inferences drawn from which various recommendations will be offered for decision making.
26
3.4 Time Series Analysis
Time series analysis involves strategies or procedures that breakdown of series into parts and
reasonable segments that enables patterns or trends to be identified, estimates and forecast to
be made. Essentially time series analysis endeavours to comprehend the hidden setting of the
information focuses using a model to estimate future esteems in light of known past esteems.
Such time series models incorporate MA, AR, ARIMA, GARCH, TARCH, EGARCH,
FIGARCH, CGARCH and ARIMA among others however the fundamental concentration of
this analysis depends on MA, AR, ARMA, and ARIMA models.
3.4.1 Lag
Lag refers to the time periods between two different observations. For instance, lag 1 is
between the times Yt and Yt-1. Lag 2 is between Yt and Yt-2. Time series can likewise be
lagged forward, Yt and Yt+1. The observation at the present time, Yt , relies upon the
estimation of the past observation, Yt-1.
3.4.2 Differencing
Differencing essentially implies subtracting the estimation of a prior observation from the
estimation of a later observation. Ascertaining contrasts among sets of observations at some
lag to make a non-stationary series stationary. There are conceivable moves in both the mean
and the scattering after some time for this series. The mean might edge upwards, and the
inconstancy might increment. On the off chance that the mean is changing, the pattern is
removed by differencing the data. In the event that the inconstancy is changing, the procedure
might be made stationary by logarithmic change. Differencing the scores is the simplest
method to make a non-stationary mean stationary (level). The quantity of times you need to
distinction the scores to make the procedure stationary decides the estimation of d. On the off
27
chance that dC0, the model is as of now stationary and has no pattern. At the point when the
series is differenced once, dC1 and direct pattern is removed. At the point when the
distinction is then differenced, dC2 and both linear and quadratic pattern are removed. For
non-stationary series, d estimations of 1 or 2 are typically sufficient to make the mean
stationary.
3.4.3 Stationary and Non-Stationary Series
Stationary series fluctuate around a consistent mean level, neither diminishing nor expanding
deliberately after some time, with steady change. Non-stationary series have efficient
patterns, for example, direct, quadratic, et cetera. A non-stationary series that can be made
stationary by differencing is called "non-stationary in the homogenous sense." Stationarity is
utilized as an instrument in time series analysis, where the crude information is regularly
changed to end up plainly stationary. For instance, monetary information is regularly
occasional or dependent on a non-stationary value level. Utilizing non-stationary time series
produces questionable and spurious outcomes and prompts poor comprehension and
anticipating. The answer for the issue is to change the time series data with the goal that it
ends up noticeably stationary. On the off chance that the non-stationary process is an
irregular stroll with or without a float, it is changed to stationary process by differencing.
Differencing the scores is the least demanding approach to make a non-stationary mean
stationary (level). The number of times you need to difference the scores to make the
procedure stationary decides the estimation of d. In the event that d=0, the model is as of now
stationary and has no pattern. At the point when the series is differenced once, d=1 and linear
pattern is expelled. At the point when the distinction is then differenced, d=2 and both linear
and quadratic pattern are expelled. For non-stationary series, d estimations of 1 or 2 are
typically satisfactory to make the mean stationary. On the off chance that the time series
information investigated shows a deterministic pattern, the spurious outcomes can be stayed
28
away from by detrending. Now and then the non-stationary series may join a stochastic and
deterministic pattern in the meantime and to abstain from acquiring deceiving comes about
both differencing and detrending ought to be connected, as differencing will evacuate the
pattern in the fluctuation and detrending will expel the deterministic pattern. A non-stationary
process with a deterministic pattern ends up noticeably stationary in the wake of expelling the
pattern, or detrending. For instance, Yt = α + βt + εt is changed into a stationary procedure by
subtracting the pattern βt: Yt - βt = α +εt. No observation is lost when detrending is utilized to
change a non-stationary procedure to a stationary one. Non-stationary information, when in
doubt, are flighty and can't be demonstrated or estimated. The outcomes got by utilizing non-
stationary time series might be spurious in that they may demonstrate a correlation between
two factors where one doesn't exist. With a specific end goal to get steady, dependable
outcomes, the non-stationary information should be changed into stationary information.
Rather than the non-stationary process that has a variable difference and an imply that does
not stay close, or comes back to a long-run mean after some time, the stationary procedure
returns around a consistent long haul mean and has a steady change free of time.
3.5 Components of Time Series
An imperative step in selecting appropriate modelling and forecasting procedure to consider
the sort of data pattern showed from the time series diagrams of the time plots. The sources of
variation in terms of patterns in time series data are mostly classified into four main
components:
3.5.1 The Trend (D)
The trend is simply the underlying long-term behaviour or pattern of the data or series. The
Australian Bureau of Statistics (ABS, 2008) defined trend as the ‘long term’ movement in a
29
time series without calendar related and irregular effects, and is a reflection of the underlying
level.
3.5.2 Seasonal Variation (S)
A seasonal effect is a systematic and calendar related effect. Some examples include the
sharp escalation in most Retail series which occurs around December in response to the
Christmas period.
3.5.3 Cyclical Variations (C)
Cyclical variations are the short-term fluctuations (rises and falls) that exist in the data that
are not of a fixed period. They are usually due to unexpected or unpredictable events such as
those associated with the business cycle sharp rise in inflation or stock price etc.
3.5.4 Irregular Variations (I)
The irregular component (sometimes also known as the residual) is what remains after the
seasonal and trend components of a time series have been estimated and removed. It results
from short term fluctuations in the series which are neither systematic nor predictable. In a
highly irregular series, these fluctuations can dominate movements, which will mask the
trend and seasonality.
3.6 Basic Assumptions in Time Series
A common assumption in many time series techniques is that the data are stationary. A
stationary process has the property that the mean, variance and autocorrelation structure do
not change over time. Stationarity can be defined in precise mathematical terms as
1. The mean µ(t)= E(γ(t))
30
2. The difference σ2(t) = Var(y(t)) = γ(0)
3. The autocovariance 𝛾(𝑡1 , 𝑡2 ) = 𝑐𝑜𝑣[𝑦(𝑡1 ), 𝑦(𝑡2 )]
Hence a time series is said to be entirely stationary if the joint distribution of any series of n
observations (𝑡1, 𝑡2 ) = cov(y(t1),y(t2)) is the same as the joint distribution of y(t1), y(t2),
y(t3)…… y(tn) for all n and k. Provided the time series isn't stationary, we can change it to
stationary with one of the methods applicable.
3.6.1 Autocorrelation Function (ACF)
4 Autocorrelation Function (ACF)
In the context of time series analysis, the relationships between observations in different time
periods play a very important role. These relationships across time can be captured by the
time series correlation respectively of covariance, known as autocorrelations resp. -
covariance.
𝛾𝑘 = 𝐸{[𝑋_𝑡 – 𝐸(𝑋_𝑡 )][𝑋_(𝑡 − 𝑘) – 𝐸(𝑋_(𝑡 − 𝑘) )]} , Where 𝑋𝑡 stands for the time-
series.
The graph of this function is called correlogram. The correlogram has an essential
importance for the analysis, because it comprised time dependence of the observed series.
Since 𝛾k and ρk only differ in the constant factor 𝛾o i.e. the autovariance of the time-series, it
is sufficient to plot just one of these two functions. One application of autocorrelation plots is
for checking the randomness in the data set. The idea is, that if these autocorrelations are near
zero for any and all time lags then the data set is random. Another application of this
correlogram is for identifying the order of an AR and an MA process. Technically, these
described plots are formed by displaying on the vertical axis the autocorrelation coefficients
(𝛾k) and on the horizontal axis, the time lag.
31
3.6.2 Partial Autocorrelation Function
The partial autocorrelation function (𝜋𝐾), where k ≥ 2, is defined as the partial correlation
between 𝑋𝑡 and 𝑋𝑡−𝑘 under holding the random variables in between 𝑋𝑢 , where t − k < u
> t , constant. It seems to be obvious, that the PACF is only defined for lags equal to two or
greater, because considering the following example: if one calculates 𝜋 2 of Xt and 𝑋𝑡−2
under holding 𝑋𝑡−1 constant then the correlation of 𝑋𝑡−1 disappears. But if one wants to
calculate the 𝜋 2 of 𝑋𝑡 and 𝑋𝑡−1 it is the same as computing the ACF at lag one, i.e. 𝜌1. The
partial autocorrelation plot or partial correlogram is also commonly used for model
identification in Box and Jenkins models. On the y-axis they display the partial
autocorrelations coefficients at lag k and on the x-axis the time lag.
32
CHAPTER FOUR
DATA ANALYSIS AND FINDINGS
This chapter contains the data analysis and presentation of results. This chapter dealt with the
trend analysis of the data, it further included time series analysis, model estimation and
selection and the forecasting for the subsequent years.
4.1 Preliminary time series analysis
In other to have a robust time series analysis, it is necessary that sufficient or lengthy time
series data is gathered, even though there is no universally accepted number of required
observations. Meyler et al., (1988) recommended a minimum of 50 observations as an
appropriate time series data for a univariate time series. It must be emphasized that the use of
few data points or observations could be problematic. On the order hand a bulky or lengthy
time series data could contains a structural break which may necessitate only examining a
sub-section of the entire data series or introduction of dummy variables. In the attempt to
overcome, the conflict between the required observations for sufficient degrees of freedom
for statistical robustness and having a few data points which devoid structural breaks. It is
imperative the data be plotted to ascertain the presence or otherwise structural breaks, outliers
within the data set. Figure 4.1: below shows Time Series plot of Monthly Malaria cases from
January 2010 to December 2017.
33
4.2 Trend of Malaria
The figure 4.1 below shows the accuracy measures of which linear trend model appears to
have accuracy measures with a Mean Absolute Error (MAPE) of 55, minimum Mean
Absolute Deviation (MAD) of 8812 and Mean Squared Deviation (MSD) of 122711204.
Therefore the data follows a linear decreasing trend model since it has the maximum
measures of accuracy. The results give an indication that the data could be useful for time
series analysis. The equation of the linear trend model is 𝑌𝑡 = 24019 − 109.0 × 𝑡
Figure 4.1 Trend analysis plot for Malaria Cases from January 2010 to December 2017
4.3 Moving Average Plot of Malaria cases in New Juaben Municipality
The Figure 4.2 also shows the corresponding moving average of the monthly malaria cases in
new juaben municipality using the same data points from january 2010 to december 2017. It
clearly shows that the malaria cases was quiet down at a certain point in time but has grown
staedily from time to time. This means that the number of people been admitted into the
34
hospital keep increasing from year to year looking at moving average graph. But finally
shows downwards behaviour for 2016 and 2017 continuosly This could be as a result of the
steady decrease in the corresponding admitted population growth over the period under
consideration.
Figure 4.2 moving average plot of malaria cases in New Juaben Municipality
4.4 Model Checking (Parameter Determination)
Autocorrelation Function for MALARIA

(with 5% significance limits for the autocorrelations)
1.0
0.8
0.6
0.4
Autocorrelation
0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0
2 4 6 8 10 12 14 16 18 20 22 24 26
Lag
Figure 4.3 Autocorrelation Function:
35
The Autocorrelation function for the malaria cases shows a significant spike at lags 1, 2, 3, 4
and 13 and all the remaining lags of Autocorrelation spikes falling within the 95% confidence
limit. Also the various test statistics in absolute value were less than 2 with the exception of
lag 1, 2, 3, 4, 5 and 13.
Partial Autocorrelation Function for MALARIA

(with 5% significance limits for the partial autocorrelations)
1.0
0.8
0.6
Partial Autocorrelation
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0
2 4 6 8 10 12 14 16 18 20 22 24 26
Lag
Figure 4.4. Partial Autocorrelation Function
The Partial Autocorrelation function for the data shows a significant spike at lags 1, 2, 3, 4
and 13 and all the remaining lags of Autocorrelation spikes falling within the 95% confidence
limit. Also the various test statistics in absolute value were less than 2 with the exception of
lag 1, 2, 3, 4, 5 and 13.
36
4.5 Model Estimation and Selection
Table 4.1: The mode of AIC and BIC value
Model Parameters AIC BIC
1. ARIMA (1, 1, 0) 12.632 13.999
2. ARIMA (0, 1, 1) 13.200 14.000
3. ARIMA (1, 1, 1) 14.286 15.122
After inputting the data in Minitab, the auto-ARIMA function in the forecast package was used. The
ARIMA model (1, 1, 0) produced the least AIC value of 12.632 and least BIC value of 13.999, hence
the best model that fits the data set and can be used for forecasting the monthly malaria cases.
The order ARIMA (P,d,q) is obtained from the combination of the order of the
Autoregressive ,number of integrated or differencing and that of moving Average process.
Five different types of ARIMA (P,d.q) models were obtained where P is the number of order
of the Autoregressive, d is the number of times the model is integrated or differenced and q
is the number of order of the moving Average. The purpose of differencing was as a result of
the original data not being stationary .The models obtained were ARIMA(1,1,0)
ARIMA (0,1,1) and ARIMA (1,1,1)
4.5 Model Selection
With the model to be selected, the mean square errors of the estimated models were
compared and the model with the least mean square error was selected. With this, the
ARIMA (1, 1, 0) model was selected as the best fit model. The general equation is written
37
as 𝑌𝑡= 𝑌𝑡−1 + 𝜑(𝑌𝑡−1 − 𝑌𝑡−2 ) − ∅1 𝑒𝑡−1 − ∅2 𝑒𝑡−1, where 𝑌𝑡 is the maternal mortality, 𝜑 is the
coefficient of the autoregressive with order 1, ∅1 is the coefficient of moving average with
order 1, ∅2 is the coefficient of moving average with order 2 and 𝑌𝑡−1 is the fitted value for
the model.
4.6 Forecasting
Forecasting plays an important role in decision making process. It is a planning tool which
helps decision makers to foresee the future uncertainty based on the behaviour of past and
current observations.
Forecasting as describe by Box and Jenkins (1976), provide basis for economic and
business planning, inventory and production control and control and optimisation of
industrial processes.
Forecasting is the process of predicting some unknown quantities. From previous studies,
most research work has found that the selected model is not necessary the model that
provides best forecasting. In this sense, further forecasting accuracy test such as ME, RMSE
and MAE must be performed on the model. Table 4.6 shows the forecasting of malaria cases
for the 12 months period with the likelihood of increase in malaria in the subsequent years.
Table 4.6 Forecast of malaria cases for 2018
Forecasts from period 96
95% Limits
Period Forecast Lower Upper Actual
JAN 10590.5 -236.6 21417.6
FEB 10601.1 -3905.3 25107.6
38
MAR 10592.9 -6903.3 28089.2
APR 10586.8 -9451.4 30625.0
MAY 10580.4 -11712.4 32873.2
JUN 10574.1 -13765.3 34913.4
JUL 10567.7 -15659.0 36794.5
AUG 10561.4 -17425.8 38548.5
SEPT 10555.0 -19088.2 40198.2
OCT 10548.6 -20662.8 41760.1
NOV 10542.3 -22162.3 43246.9
DEC 10535.9 -23596.6 44668.5
Interpretation: the result above is forecast value and the interval in which the forecast values
lies. This could be seen that in the month of January the lower forecast value would be
8052.5 -236.6 and highest forecast value would be 8052.5+ 21417.6 in that order for all the
months.
39
CHAPTER FIVE
DISCUSSIONS, CONCLUSIONS AND RECOMMENDATIONS
This chapter discusses the findings of the study, draws relevant conclusions based on these
results and appropriate recommendations made with a view to improving malaria infection in
Ghana.
5.1 Findings
The general objective of the study was to statistically analyse Malaria cases within the New
Juaben Municipality. In order to arrive at this main objective, the researcher was guided by
three specific objectives: Develop appropriate time series model for malaria cases in the New
Juaben Municipality, determine the trend of malaria cases in the New Juaben Municipality
from 2010-2017, and to forecast malaria cases in the Municipality for the next 12 month
period (i.e. from January, 2018 to December, 2018).
The findings of the results are therefore presented in line with these objectives. In order to
identify the appropriate time series model for malaria cases within the New Juaben
Municipal, several time series models including AR, MA, ARMA, non-seasonal
Autoregressive Integrated Moving Average (ARIMA) and seasonal ARIMA were used in
modelling the data in the Minitab14.
The study identified several ‘candidate’ models which best fitted the data. However, with the
use of the Modified Box-Pierce (Ljung-Box) Chi-square statistic criteria of the “largest p-
value and minimum Chi-Square value,” the best-fitted ARIMA model selected was ARIMA
(1, 1, 0). After the estimation of the parameters of selected models, a series of diagnostic and
forecasting accuracy tests were performed. The general equation is written as 𝑌𝑡= 𝑌𝑡−1 +
𝜑(𝑌𝑡−1 − 𝑌𝑡−2 ) − ∅1 𝑒𝑡−1 − ∅2 𝑒𝑡−1 , where 𝑌𝑡 is the maternal mortality, 𝜑 is the coefficient
of the autoregressive with order 1, ∅1 is the coefficient of moving average with order 1, ∅2 is
40
the coefficient of moving average with order 2 and 𝑌𝑡−1 is the fitted value for the model. The
resulting model is 𝑌𝑡 = 𝑌𝑡−1 + 0.3521(𝑌𝑡−1 − 𝑌𝑡−2 ) − 0.8821𝑒𝑡−1 − 0.1556𝑒𝑡−1 .
5.2 Conclusions
With reference to the findings of the research, it can be concluded that:
The most adequate model for the data was ARIMA (1, 1, 0) and also the results of the
forecast shows malaria showed a decreasing trend in the municipality. Finally the model was
used to forecast future malaria cases for the following 12 months period. The results of the
forecasts revealed a decreasing trend of malaria cases within the municipality.
The forecasted malaria cases in the municipality from January, 2018 to December, 2018 were
respectively 8053, 5491, 4357, 3726,3271, 2878, 2508, 1785, 1426, 1067 and 709.
5.2 Recommendations
On the basis of the findings of the research, the following recommendations were made:
(i) The Ghana Health Service authorities in Ghana should use the ARIMA (1, 1, 0) model
in determining malaria cases. This may also be adopted by regional and municipal health
directorates and other health institutions in Ghana.
(ii) The predicted malaria cases using the above model could greatly help the Government
of Ghana, Health institutions and regulators of health systems in its operational activities.
(iii) Government through the Ministry of Health should put in effective and efficient
malaria reduction strategies across the municipalities to help sustain or reduce the rate of
malaria infection among the citizenry.
41
REFERENCES
Akazili, J. (2002, November). Costs to households of seeking malaria care in the Kassena-
Nankana District of Northern Ghana. In Third MIM Pan-African Conference on Malaria,
Arusha, Tanzania (pp. 17-22).
Alaii, J. A., Hawley, W. A., Kolczak, M. S., Ter Kuile, F. O., Gimnig, J. E., Vulule, J. M., ...
& Phillips-Howard, P. A. (2003). Factors affecting use of permethrin-treated bed nets during
a randomized controlled trial in western Kenya. The American journal of tropical medicine
and hygiene, 68(4_suppl), 137-141.
Al‐Taiar, A., Jaffar, S., Assabri, A., Al‐Habori, M., Azazy, A., Al‐Gabri, A., ... & Whitty, C.
J. (2008). Who develops severe malaria? Impact of access to healthcare, socio‐economic and
environmental factors on children in Yemen: a case‐control study. Tropical medicine &
international health, 13(6), 762-770.
Brown, B. E. (1997). Coral bleaching: causes and consequences. Coral reefs, 16(1), S129-
S138.
Bryman, A. (2001). The nature of qualitative research. Social research methods, 365-399.
Collins, P. H. (1997). Comment on Hekman's" Truth and Method: Feminist Standpoint
Theory Revisited": Where's the Power?. Signs: Journal of Women in Culture and Society,
22(2), 375-381.
Dysoley, L., Kaneko, A., Eto, H., Mita, T., Socheat, D., Börkman, A., & Kobayakawa, T.
(2008). Changing patterns of forest malaria among the mobile adult male population in
Chumkiri District, Cambodia. Acta tropica, 106(3), 207-212.
Filmer, D. (2002). Fever and its treatment among the more and less poor in sub-Saharan
Africa. The World Bank.
Holton, E. B. (1997). M.. Quantitative Research Methods. Human Resource Development:
Lingking Research and Practice.
42
Holtz, J. (2002). Sensorless control of induction motor drives. Proceedings of the IEEE,
90(8), 1359-1394.
Krause, G., & Sauerborn, R. (2000). Comprehensive community effectiveness of health care.
A study of malaria treatment in children and adults in rural Burkina Faso. Annals of tropical
paediatrics, 20(4), 273-282.
Lines, J., Lengeler, C., Cham, K., De Savigny, D., Chimumbwa, J., Langi, P., ... & Lynch, M.
(2003). Scaling-up and sustaining insecticide-treated net coverage. The Lancet Infectious
Diseases, 3(8), 465-466.
Makemba, A. M., Winch, P. J., Kamazima, S. R., Makame, V. R., Sengo, F., Lubega, P. B.,
... & Shiff, C. J. (1995). Community-based sale, distribution and insecticide impregnation of
mosquito nets in Bagamoyo District, Tanzania. Health Policy and Planning, 10(1), 50-59.
Malaria, R. B. (2001). Scaling-up insecticide-treated netting programmes in Africa. A
strategic framework for coordinated national action. Geneva, Switzerland.
Matveev, A. V. (2002). The advantages of employing quantitative and qualitative methods in
intercultural research: Practical implications from the study of the perceptions of intercultural
communication competence by American and Russian managers. Theory of communication
and applied communication, 1(1), 59-67.
Mboera, L. E., Makundi, E. A., & Kitua, A. Y. (2007). Uncertainty in malaria control in
Tanzania: crossroads and challenges for future interventions. The American journal of
tropical medicine and hygiene, 77(6_Suppl), 112-118.
Ménard, D., Barnadas, C., Bouchier, C., Henry-Halldin, C., Gray, L. R., Ratsimbasoa, A., ...
& Bertrand, O. (2010). Plasmodium vivax clinical malaria is commonly observed in Duffy-
negative Malagasy people. Proceedings of the National Academy of Sciences, 107(13), 5967-
5971.
43
Monitoring, R. B. M. (2005). Evaluation Reference Group. World Health Organization,
United Nations ChildrenLs Fund, MEASURE DHS, MEASURE Evaluation, and US Centers
for Disease Control and Prevention: Malaria Indicator Survey: Basic documentation for
survey design and implementation Calverton, Maryland: MEASURE Evaluation.
Nayyar, G. M., Breman, J. G., Newton, P. N., & Herrington, J. (2012). Poor-quality
antimalarial drugs in southeast Asia and sub-Saharan Africa. The Lancet infectious diseases,
12(6), 488-496.
Nayyar, G. M., Breman, J. G., Newton, P. N., & Herrington, J. (2012). Poor-quality
antimalarial drugs in southeast Asia and sub-Saharan Africa. The Lancet infectious diseases,
12(6), 488-496.
Nganda, R. Y. (2004). Knowledge of malaria influences the use of insecticide treated nets but
not intermittent presumptive treatment by pregnant women in Tanzania. Malaria Journal,
3(1), 42.
Nganda, R. Y., Drakeley, C., Reyburn, H., & Marchant, T. (2004). Knowledge of malaria
influences the use of insecticide treated nets but not intermittent presumptive treatment by
pregnant women in Tanzania. Malaria Journal, 3(1), 42.
Nuwaha, F. (2001). Factors influencing the use of bed nets in Mbarara municipality of
Uganda. The American journal of tropical medicine and hygiene, 65(6), 877-882.
Oberländer, L., & Elverdan, B. (2000). Malaria in the United Republic of Tanzania: cultural
considerations and health-seeking behaviour. Bulletin of the World Health Organization, 78,
1352-1357.
Rowland, I. R., Wiseman, H., Sanders, T. A., Adlercreutz, H., & Bowey, E. A. (2000).
Interindividual variation in metabolism of soy isoflavones and lignans: influence of habitual
diet on equol production by the gut microflora. Nutrition and cancer, 36(1), 27-32.
44
Schapira, A., Beales, P. F., & Halloran, M. E. (1993). Malaria: living with drug resistance.
Parasitology today, 9(5), 168-174.
Schlaepfer, T. E., Cohen, M. X., Frick, C., Kosel, M., Brodesser, D., Axmacher, N., ... &
Sturm, V. (2008). Deep brain stimulation to reward circuitry alleviates anhedonia in
refractory major depression. Neuropsychopharmacology, 33(2), 368.
Schönfeld, M., Miranda, I. B., Schunk, M., Maduhu, I., Maboko, L., Hoelscher, M., ... &
Löscher, T. (2007). Molecular surveillance of drug-resistance associated mutations of
Plasmodium falciparum in south-west Tanzania. Malaria Journal, 6(1), 2.
Sekaran, U., & Bougie, R. (2011). Business Research Methods: A skill-building approach.
Snover, K. E. (2005). Perceptions of malaria among Western populations and the Wandamba
of Eastern Africa: A cross-cultural comparison.
Snow, R. W., De Azevedo, I. B., Lowe, B. S., Kabiru, E. W., Nevill, C. G., Mwankusye, S.,
... & Teuscher, T. (1994). Severe childhood malaria in two areas of markedly different
falciparum transmission in East Africa. Acta tropica, 57(4), 289-300.
Teklehaimanot, A., Sachs, J. D., & Curtis, C. (2007). Malaria control needs mass distribution
of insecticidal bednets. The Lancet, 369(9580), 2143-2146.
Trochim, W. M. (2006). Qualitative validity.
Ukoli, P. (1990). Q fever in Sokoto, Nigeria. Annals of the New York Academy of Sciences,
590(1), 281-282.
Unicef. (2003). The Africa malaria report 2003 (No. WHO/CDS/MAL/2003.1093). Geneva:
World Health Organization.
Vermund, S. H., & Yamamoto, N. (2007). Co-infection with human immunodeficiency virus
and tuberculosis in Asia. Tuberculosis, 87, S18-S25.
Victora, C. G.(2003). Applying an equity lens to child health and mortality: more of the same
is not enough. The Lancet, 362(9379), 233-241.
45
Wandiga, S. O., Opondo, M., Olago, D., Githeko, A., Githui, F., Marshall, M., ... &
Kabumbuli, R. (2006). Vulnerability to climate induced highland malaria in East Africa.
AF91 Synthesis Paper-submitted to Climate Change.
World Health Organization. (2003). Diet, nutrition, and the prevention of chronic diseases:
report of a joint WHO/FAO expert consultation (Vol. 916). World Health Organization.
Yadav, G. D., & Nair, J. J. (1999). Sulfated zirconia and its modified versions as promising
catalysts for industrial processes. Microporous and mesoporous materials, 33(1-3), 1-48.
Ziba, C., Slutsker, L., Chitsulo, L., & Steketee, R. W. (1994). Use of malaria prevention
measures in Malawian households. Tropical medicine and parasitology: official organ of
Deutsche Tropenmedizinische Gesellschaft and of Deutsche Gesellschaft fur Technische
Zusammenarbeit (GTZ), 45(1), 70-73.
46
APPENDIX
Monthly Malaria cases from 2010-2017
Male Female Grand
Total Total Total
2010
JAN 4218 7196 11414
FEB 3801 5579 9380
MARCH 4118 6436 10554
APRIL 4474 6981 11455
MAY 4701 7218 11919
JUNE 5924 8630 14554
JULY 6046 8782 14828
AUG 4229 6698 10927
SEP 4099 6429 10528
OCT 5054 7275 12329
NOV 5061 7938 12999
DEC 3869 6158 10027
2011
JAN 3615 5830 9445
FEB 3394 5444 8838
MARCH 4959 7690 12649
APRIL 3922 6217 10139
MAY 4426 6558 10984
47
JUNE 4993 7475 12468
JULY 4161 6411 10572
AUG 4055 6650 10705
SEP 16817 20665 20872
OCT 18402 34987 35219
NOV 19348 37526 37750
DEC 17146 36330 36494
2012
JAN 17288 17081 17288
FEB 15080 32170 32368
MARCH 14667 29587 29747
APRIL 16065 30561 30732
MAY 18477 34263 34542
JUNE 21468 39664 39945
JULY 23546 44690 45014
AUG 21065 44372 44611
SEP 18849 39683 39914
OCT 19446 38004 38295
NOV 16663 35784 36109
DEC 6151 22714 22814
2013
JAN 18288 18014 18288
FEB 14838 32900 33126
MARCH 11867 26526 26705
48
APRIL 16611 28279 28478
MAY 18258 34665 34869
JUNE 21603 39494 39861
JULY 22401 43735 44004
AUG 24350 46439 46751
SEP 20489 44570 44839
OCT 21453 41620 41942
NOV 17570 38801 39023
DEC 19120 35928 36690
2014
JAN 8759 8639 8759
FEB 9072 17672 17831
MARCH 8423 17367 17495
APRIL 9952 18211 18375
MAY 7007 16820 16959
JUNE 6650 13547 13657
JULY 6439 12980 13089
AUG 7131 13454 13570
SEP 5438 12491 12569
OCT 5410 10755 10848
NOV 7079 12374 12489
DEC 4822 11824 11901
2015
JAN 5602 5524 5602
49
FEB 4987 10521 10589
MARCH 5577 10470 10564
APRIL 7621 13085 13198
MAY 7257 14715 14878
JUNE 5820 12972 13077
JULY 7673 13343 13493
AUG 8492 16042 16165
SEP 7321 15682 15813
OCT 6134 13338 13455
NOV 6610 12655 12744
DEC 6967 13497 13577
2016
JAN 4559 4497 4559
FEB 3079 7598 7638
MARCH 3620 6669 6699
APRIL 7707 11218 11327
MAY 8607 16153 16314
JUNE 10126 18550 18733
JULY 8221 18219 18347
AUG 7933 16052 16154
SEP 6720 14566 14653
OCT 5856 12469 12576
NOV 5761 11519 11617
DEC 4952 10631 10713
50
2017
JAN 6737 6605 6737
FEB 5240 11897 11977
MARCH 4332 9506 9572
APRIL 4172 8462 8504
MAY 5709 9776 9881
JUNE 6103 11691 11812
JULY 5571 11587 11674
AUG 7329 12777 12900
SEP 5237 12498 12566
OCT 4156 9312 9393
NOV 5153 9245 9309
DEC 5601 10661 10754
AUTOCORRELATION (ACF)
Lag PACF T
1 0.725246 7.36
2 0.158019 1.60
3 0.225163 2.29
4 0.073082 0.74
5 -0.092600 -0.94
6 -0.080285 -0.81
7 0.004961 0.05
8 0.104227 1.06
9 0.239205 2.43
51
10 0.134592 1.37
11 -0.041463 -0.42
12 0.068264 0.69
13 0.375528 3.81
14 -0.406782 -4.13
15 -0.018035 -0.18
PARTIA AUTOCORRELATION (PACF)
Lag ACF T LBQ
1 0.725246 7.36 55.77
2 0.600886 4.26 94.43
3 0.585596 3.57 131.52
4 0.544452 2.97 163.90
5 0.437911 2.21 185.06
6 0.348553 1.68 198.61
7 0.313686 1.47 209.70
8 0.318155 1.46 221.22
9 0.372248 1.68 237.16
10 0.390588 1.71 254.90
11 0.353751 1.51 269.61
12 0.394047 1.64 288.07
13 0.569320 2.32 327.02
14 0.414601 1.61 347.90
15 0.340687 1.29 362.17
52

Thesis-Draft2 - Copy Final

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Thesis-Draft2 - Copy Final

Transféré par

Droits d'auteur :

Formats disponibles

KOFORIDUA TECHNICAL UNIVERSITY

FACULTY OF APPLIED SCIENCE AND TECHNOLOGY

DEPARTMENT OF APPLIED MATHEMATICS

TIME SERIES ANALYSIS OF MALARIA IN THE NEW JUABEN MUNICIPALITY

A PROJECT REPORT PRESENTED TO THE FACULTY OF APPLIED SCIENCE

AND TECHNOLOGY, DEPARTMENT OF APPLIED MATHEMATICS,

KOFORIDUA TECHNICAL UNIVERSITY IN PARTIAL FULFILLMENT OF THE

REQUIREMENT FOR THE AWARD OF BACHELOR OF TECHNOLOGY IN

of Technology in Statistics and that, to the best of my knowledge, It contains no material

made in the text.

and regulations governing project work supervision in Koforidua Technical University. I

Signature: …………..……….. Date………………

APPIAH ANDANE AKABOHA

I again respectively dedicate it to my lovely families and loved ones.

I also wish to express my deepest appreciation to my supervisor Mr. Appiah Andane

maternal mortality, 𝜑 is the coefficient of the autoregressive with order 1, ∅1 is the

LIST OF TABLES viii

Table 4.1: The mode of AIC value

1.1 Background of the study

Malaria is a female-mosquito-borne infectious disease caused by a eukaryotic protist of the

productivity and economic loss.

inadequate mosquito-control measures such as spraying insecticides inside houses and

draining standing water where mosquitoes lay their eggs.

increasing trend of the disease.

the disease accounts for 20% of all childhood deaths.

(Sudhakar et at., 2007).

continent and in the poorest countries (Worral et al., 2003).

Health Service to allocate adequate resources to efficiently control the disease.

1.2 Problem statement

launched of the Nationwide Malaria Control Programme by Zoomlion Ghana Limited at

rate of malaria in New Juaben municipality and Ghana as a whole.

1.3 General Objective

municipality of eastern region of the Republic of Ghana.

1.3.1 Specific Objectives

1.4 Research Questions

1.5 Significance of the study

have also joined in the fight against the disease.

governmental organizations and other health-care services to respond to outbreak in timely

other non-governmental organizations to help control the disease.

national expenditure budgets on health care.

to literature already in existence.

1.7 Organization of the Study

method of data analysis technique.

as well as interpretation of the results.

The final chapter will engage on discussions, conclusions and recommendations.

2.1 The Nature of Malaria

From the study of Nayyar, et al (2006), Malaria is a mosquito-borne infectious disease of

humans and other animals caused by parasitic protozoans(a type of unicellular

and draining standing water.

ongoing. Several medications are available to prevent malaria in travellers to malaria-

endemic countries (prophylaxis). A variety of antimalarial medications are available. Severe

developed to several antimalarial drugs; for example, chloroquine-resistant P. falciparum

problem in some parts of Southeast Asia (Nayyar, et al, 2006).

2.2. Global Burden of Malaria Disease

are reasons for an increasing anaemia burden in children (Greenwood et al.1991).

2.3 Human Related Factors

&Staalsoe T.2004).The protective effect of maternal antibody is likely to be less when

the individual adults (Alaii et al, 2003).

2.4 Knowledge on Malaria

child-caring role of women and children‘s increased vulnerability to malaria. Furthermore,

that were done by

a complicated relationship since educational attainment can have the impact on an

individual‘s ability to understand and access information regarding malaria prevention