Vous êtes sur la page 1sur 5

ECON1203: Statistics

techniques

Chapter 4: Numerical descriptive

ECON1203: Statistics
Chapter 4: Numerical descriptive
techniques
1. Measures of central location
a. Arithmetic mean
N

xi

Population mean: = i=1


N
n

xi

Sample mean: x = i=1


n

When to use the mean


1. Objective: Describe a single set of data
2. Type of data: Interval
3. Descriptive measurement: Central location

b. Median
Location of m edian=

n+1
2

When to use the median


1. Objective: Describe a single set of data
2. Type of data: Ordinal or interval (with extreme observations)
3. Descriptive measurement: Central location

c. Mode
Mean=Observation with highest frequency
When to use the mode
1. Objective: Describe a single set of data
2. Type of data: Nominal, ordinal, interval

ECON1203: Statistics
techniques

Chapter 4: Numerical descriptive

d. Geometric mean
Geometric mean=

i=1

( 1+ r i)1

When to use the geometric mean


1. Objective: Describe a single set of data
2. Type of data: Interval; growth rates

2. Measures of variability
When to use the range, variance, standard deviation, and coefficient of
variation
1. Objective: Describe a single set of data
2. Type of data: Interval
3. Descriptive measurement: Variability

a. Range
Range=Largest observationSmallest observation

b. Variance
N

( x i )2

Population variance : 2= i=1

n
2

Sample variance :s =

( x ix )
i=1

n1

1
=
n1

2
i

x
i=1

( )
xi

c. Standard deviation
Population standard deviation: = 2
Sample standard deviation: s= s2

i=1

ECON1203: Statistics
techniques

Chapter 4: Numerical descriptive

d. Empirical Rule
1. Approximately 68% of all observations fall within one standard deviation of
the mean.
2. Approximately 95% of all observations fall within two standard deviation of
the mean.
3. Approximately 99.7% of all observations fall within three standard
deviation of the mean.

e. Chebysheffs Theorem
The proportion of observations in any sample or population that lie within
standard deviations of the mean is at least

f.

1
[ for k >1 ]
2
k

Coefficient of variation

Population coefficient of variation : CV =

Sample coefficient of variation : cv=

s
x

3. Measures of relative standing


a. Location of a percentile
LP =( n+1 )

P
100

b. Interquartile range
Interquartile range=Q3Q1

c. Box plot

ECON1203: Statistics
techniques

Chapter 4: Numerical descriptive

4. Measures of linear relationship


When to use covariance, coefficient of correlation, coefficient of
determination, and least sqs line
1. Objective: Describe the relationship between two variables
2. Type of data: Interval

a. Covariance

Covariance is positive.
A large covariance indicates a strong linear relationship.
A small covariance indicates a weak linear relationship.
N

( x i x ) ( y i y )

Population covariance : xy = i=1

( xi x ) ( y i y )
Sample covariance :s xy =

i=1

n1

1
=
n1

xi y i

i =1

x i y i i=1

i=1

b. Coefficient of correlation

A positive coefficient of correlation indicates a positive linear relationship.


A negative coefficient of correlation indicates a negative linear
relationship.
A large coefficient of correlation indicates a strong linear relationship.
A small coefficient of correlation indicates a weak linear relationship.

Population coefficient of correlation : =

Sample coefficient of correlation :r =

xy
x y

s xy
sx s y

c. Coefficient of determination

Perfect positive linear relationship: R2=1

Nolinear relationship: R 2=0

Perfect negative linear relationship : R2=1

ECON1203: Statistics
techniques

Chapter 4: Numerical descriptive

Coefficient of determination : R2=r 2

d. Least squares method


( ^y =b 0 +b1 x)

The least squares line

[ (
n

deviations

b1=

i=1

y i^y i )2

minimises the sum of the squared

s xy
s2x

b0 = y b 1 x

Vous aimerez peut-être aussi