Measures of Central Tendency Measures of central Value are simple statistical treatments of distribution that attempts to find the

single figure to describe the entire distribution. It is the best possible value of a group of variables that singly represents to whole group. In the statistical analysis the central value falls within the approximately middle value of the whole data. Among the several tools of measuring central value, the mean has been used in this analysis where and when necessary. The mean is the arithmetic average of a variable. Arithmetic Mean of a series given by: Mean () =

Where, X=Sum of the variable 'X' and N= No. of observation Measure of Dispersion Dispersion measures the variation of the data from the central value. The central value alone is not enough to analyze the quality of data regarding its variability. With the light of dispersion, an average becomes more powerful and meaningful. Following tools of measuring dispersion has been used in this study.

Standard Deviation Standard Deviation (S.D) is the most popular and the most useful measure of dispersion. It includes the ranges and size of deviance from the middle or mean. It measures the absolute dispersion. Higher the value of standard deviation higher is the variability and vice versa. It is the positive square root of average sum of squares of deviations of observations from the arithmetic mean of the distribution. It can be calculated as follows: Standard Deviation ( )=


Co-efficient of Variation The percentage measure of coefficient of standard deviation is called coefficient of variation. The less is the C.V. the more is the uniformity and consistency and vice versa. Standard deviation gives an absolute measure of dispersion. Hence where the mean value of the variable

is not equal it is not appropriate to compare two pairs of variables based in S.D. only. The coefficient of variation measures the relative measures of dispersion, hence capable to compare two variables independently in terms of their variability.

    

  

Correlation Coefficient (r) Correlation refers to the degree of relationship between two variables. Correlation coefficient determines the association between the dependent variable and independent variable. One variable is treated as dependent and another variable is treated as independent. In between the variables, increase or decrease in one cause increase or decrease in another, then such variables are correlated variables. "Correlation may be defined as the degree of linear relationship existing between two or more variables. Two variables are said to be correlated if the change in the value of one variable appears to be related or linked with the change in the other variables. It refers to closeness of the relationship between two or more variables. Correlation says just degree of relationship between two or more variables. It does not tell us anything about cause and effect relationship i.e. if there is a high degree of correlation between two variables we cannot say which the cause is and which is the effect. There are different techniques of calculating correlation coefficient. Among various techniques we have used Karl Pearson coefficient of correlation. The Karl Pearson Coefficient always falls between -1 to +1.The value of correlation of coefficient in -1 signifies the negative correlation and in +1 signified the positive correlation coefficient. It is calculated as follows: Correlation Coefficient (r) =

 

Where, x=X-
x=Standard

y=Y- Deviation of Series X

y=

Standard Deviation of Series Y

N= No. of pairs of observation If, r=0, there is no relationship between the variables r<0, there is negative relationship between the variables r>0, there is positive relationship between the variables r=-1, the relationship is perfectly negative between the variables

r=+1, the relationship is perfectly positive between the variables

The correlation coefficient gives the actual relationship but sometimes it may give the error. The reliability of the correlation coefficient (r) can be checked with the help of probable error (PE).

Probable Error (P.E) of Correlation Coefficient The probable error is a measurement of ascertaining the reliability of the value of coefficient of correlation. It is used to test whether the calculated value of sample correlation coefficient is significant or not. If r is the calculated correlation coefficient in a sample of n pairs of observations, then its standard error, usually denoted by S.E (r) is given by:

S.E (r) =

Probable error of the coefficient of correlation can be also be calculated from S.E of the coefficient of correlation by the following formula

Probable Error P.E (r ) =

PE =

 

Where, r= correlation coefficient n= no of observation

A few rules for the interpretation of the significance of correlation coefficient are as follows: Decision: i. ii. iii. If r> 6 PE, then correlation coefficient is significant and reliable. If r< 6 PE, then the correlation coefficient is insignificant and there is no evidence of correlation. In other situations, nothing can be calculated with certainty.

Correlation Analysis The percentage measure of coefficient of standard deviation is called coefficient of variation. The less is the C.V then more is the uniformity and consistency and vice versa. Standard deviation gives an absolute measure of dispersion. Hence where the mean of the variable is not equal it is not appropriate to compare two pairs of variables based in Standard Deviation only. The coefficient of variation measures the relative measures of dispersion, hence capable to compare two variables independently in term of their variability.

Correlation between Loans and Advances and Total Deposit This correlation between the loans and advances and total deposits describes the degree of relationship between these two variables. Deposit is one of the major items of liability side and loan and advances is the major item of assets side of balance sheet of any commercial banks. Banks disburse loans and advances through the funds received from the depositors. In this case, the deposit is the independent variable and loan and advances is the dependent variable. It shows how a unit increase in deposit impact in the volume of the loan and advances is exhibited by this correlation coefficient.

Correlation between Loans and Advances and Total Deposits

Year 2007 2008 2009 2010 2011

Loans and Advances(X) 15,545,778 21,365,053 27,589,933 32,268,873 38,034,097

Total Deposit(Y) 23,342,285 31,915,047 37,348,255 49,696,112 46,410,700

x=X- -11,414,969 -5,595,694 629,186 5,308,126 11,073,350

y=Y- -14,400,195 -5,827,433 -394,225 11,953,632 8,668,220

xy 164377774355922 32608529588877 -248040803858 63451388265984 95986237885314 xy= 356175889292239

130301512704973 31311789103358 395875274270 28176203755126 122619084651840 2 x = 312804465489569

207365610277947 33958973038516 155413192935 142889322772877 75138041435688 2 y= 459507360717963

The amount is in thousands

Mean =26960747185 S. D. r= PE=

x


y

37742480439 =9586525.551

=7909544.43

0.939467604 0.03541338 6 P.E= 0.212480282

Here the Karl Pearson Coefficient has the value of .9394 is greater than 6 P.E. (=0.212480282), therefore the correlation coefficient is relevant and reliable. The positive value shows that loans and advances and the total deposits are positively correlated

Correlation between Loan Loss Provision and Loans and Advances The correlation between Loan Loss Provision and loans and advances shows the degree of relationship between these two items. How a unit increment in loans and advances affect the loan loss provision is measured by this correlation. Here loans and advances is independent variable and Loan Loss Provision is dependent variable.

Correlation between Loan Loss Provision and Loans and Advances

Year 2007 2008 2009 2010 2011

Loans loss Provision(X) 357,245 394,407 409,079 762,095 871,390

Loans and Advances(Y) 15,545,778 21,365,053 27,589,933 32,268,873 38,034,097

x=X- -201,598 -164,436 -149,764 203,252 312,547

y=Y- -11,414,969 -5,595,694 629,186 5,308,126 11,073,350

xy 2301234920462 920133538584 -94229412104 1078887225752

40641753604.00 27039198096.00 22429255696.00 41311375504.00 97685627209.00 x2= 229107210109.00

130301517270961.00 31311791341636.00 395875022596.00 28176201631876.00

122619080222500.00 3460942322450 y2= xy= 312804465489569.00 7666968595144 The amount is in thousands

Mean S.D.= r= P.E. =

=558,843,216
x =214059422.7

26,960,747,185
y =7909544369

0.905665531

0.054226808

6 P.E. = 0.325360847

Here, the Karl Pearson Coefficient has the value of .9056655 is greater than 6 P.E. (.= 0.325360847),therefore the correlation coefficient is relevant and reliable. The positive value shows that loans and advances and the total deposits are positively correlated.

Correlation between Loan Loss Provision and Non-Performing Loan The correlation between Loan Loss Provision & Non-Performing Loan describes the relationship between them. How a unit increase in Non-Performing Loan effect the Loan Loss Provision is exhibited by this correlation. Here non-performing loan is independent variable and Loan Loss Provision is dependent variable. Non-Performing Loan are the loan falling on the category of substandard, Doubtful and Loss Loan and the respective provisioning requirement is 25%, 50% and 100%.Higher the NPL higher will be the provisioning amount.

Correlation between Loan Loss Provision and Non-Performing Loan

Year 2007 2008 2009 2010 2011

Loans loss Provision(X) 357,245 394,407 409,079 762,095 871,390

Non-Performing Loan(Y) 264,541 171,717 224,817 487,542 689,852

x=X- -201,598 -164,436 -149,764 203,252 312,547 y=0

y=Y- -103,153 -195,977 -142,877 119,848 322,158 x=0

Xy 10640541409 38406984529 20413837129 14363543104 20795459125 32225713167 21397859603 24359321726

40641834243 27039263870 22429315602 41311294203

97685502190 103785776964 100689451994 x2= y2= xy= 187610683135 199467805616 229107210109 The amount is in thousands

Mean SD r PE

=558,843,216
x =214059422.7

367,694,100

y =193706109.9

0.962109551 0.022425904

6 P.E=0.134555421

Here, the Karl Pearson Coefficient has the value of .962109551 is greater than 6 P.E. (=0.134555421), therefore the correlation coefficient is relevant and reliable. The positive value shows that loans and advances and the total deposits are positively correlated.

Correlation between Loans and Advances And Interest Expense

Year 2007 2008 2009 2010 2011

Loans and Advances(X) 15,545,778 21,365,053 27,589,933 32,268,873 38,034,097

Interest Expense 555,710 758,436 1,153,280 1,960,107 2,955,430

x=X-x 11,414,969 -5,595,694 629,186 5,308,126 11,073,350 0

y=Y-y -920,883 -718,157 -323,313 483,514 1,478,837 0

x2 130301512704973 31311789103358 395875274270 28176203755126 122619084651840 312804465489569

y2 848024762983 515748902124 104531037319 233786175007 2186960055639 3889050933071

xy 10511846147463 4018584434049 -203423826206 2566555454717 16375684419058 33269246629081

Mean SD r PE

=26960746.8
x =7909544.43


y =881935.477

0.95386 0.0271932627

6 P.E=0.16316176

Here, the Karl Pearson Coefficient has the value of 0.95386 is greater than 6 P.E. (=0.16316176), therefore the correlation coefficient is relevant and reliable. The positive value shows that loans and advances and the total deposits are positively correlated. The Karl Pearson Coefficient is nearly equal to one therefore they are highly related.