CHAPTER ONE

INTRODUCTION

1.1 Definition and Concept of Econometrics

– Econometrics concerns with the application of mathematical statistics, and

tools of statistical inference to the empirical measurement of relationships
postulated by economic theory (Greene,2003).

– It is the social science that applies the tools of economic theory,

mathematics and statistical inference to analyze economic phenomena and
estimate causal relationships among variables.

– Estimate causal relationships means how a change in one variable affects

the other.
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 1
 – Econometrics is the quantitative measurement and analysis of actual
economic and business phenomena which involves economic theory,
statistics, mathematics and observation/data collection.

– Three major uses of econometrics:

• Describing economic reality,
• Testing hypotheses about economic theory and
• Forecasting future economic activity

– So econometrics is all about questions: the researcher first asks

questions and then uses econometrics to answer them

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 2

 – Thus, econometrics is an amalgamation/ combination of economic
theory, mathematics and statistics of the three subjects which is vital
as illustrated below:

(a) Economic theory: forms the basis for any econometric work, and
should really be the starting point for econometric analysis.

– However, lack of empirical content in economic theory is of our

interest to use econometrics for verifying economic theory empirically.

– For example, economic theory tells us people tend to consume more in

consumption expenditure, whenever their disposable income rises.

– The theory doesn’t explain by how much consumption expenditure will

increases as income increases by certain unit.

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 3

(b) Mathematical Economics: It is used to covert the economic theory into an
equation for empirically tested with the help of econometrics

 The simplest linear economic model example is:

Y = β0 +β1 X

β0 is the constant or intercept coefficient term and β1 is the slope coefficient,

the amount that Y will change when X changes by certain amount.

(c) Statistics: It provides the econometrician with the know-how and tools of
data collection, processing, analysis and presentation of results. Thus statistics
helps to test theories and explain the results.

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 4

 1.2 Specification of Econometric Model
– In econometric theory, there is not always a perfect or exact
relationship among the variables.

– Instead, there exists an inexact or stochastic relationship among most

economic variables. Thus, by introducing the error term, relevant
econometric model can be specified as
Y  1   2 X  u

– The mathematical model/equation is not a complete description of

origins of variation in Y

– Because at least four sources of variation in Y, other than the variation

in the included Xs exists.

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 5

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 6
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 7
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 8
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 9
 Cross-sectional Data: Data on one or more variable collected at the same
point in time.
– Examples are survey on households, opinion polls conducted by various
polling organizations and temperature at a given time in several places
Time Series Data: A set of observations that a variable takes at different
times, such as
– Daily e.g., stock prices; Weekly e.g., money supply;
– Monthly e.g., unemployment rate;
– Annually/yearly e.g., GDP and government budgets;
– Decennially or every ten years e.g., the census of population.
Panel Data: Combines features of both cross-section and time series data
and Same cross-sectional units are followed over time.
– Panel data represents a special type of pooled data where the
same cross-sectional units are not necessarily followed over
time.
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 10
 (ii) Regression Analysis is formally a statistical technique that attempts to
explain movements/changes in the dependent variable as a function of
the independent or explanatory variables, through the quantification
of a single equation.

– Statistically, the significant regression result does not necessarily imply

causality because we need:
• Economic theory
• Common sense
– Regression is an econometric technique for estimating the relationships
between the change in the dependent and independent variables.

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 11

– There always a distinction made between correlation and regression.
correlation analysis aims at finding the strength of linear association between
variables,
– While regression analysis, on the other hand, aims at finding the direction of
relationship between variables.
– Thus, correlation does not imply causality, but regression does so. Thus two
primary differences between them as outlined below:

CORRELATION REGRESSION
Assumes symmetry between X and Y Assumes asymmetry between X and
i.e., there is no distinction as to Y; i.e., distinguishes which variable is
which variable is dependent/causality dependent and which is explanatory.
is not important. causality is important.

Both X and Y are assumed to be Only Y is assumed to be statistical but

statistical, random or stochastic X is assumed to be fixed.

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 12

 – Regression helps to analyse how the typical value of the dependent
variable changes when any one of the independent variables changes,
while the other are held constant/ceteris paribus.

– The main purpose of linear regression analysis is to assess associations

between dependent and independent variables in the case of multiple
regression analysis.

– There are basically two types of regression analysis, i.e., simple regression
analysis and multiple regression analysis.

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 13

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 14
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 15
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 16
1.4 Assumptions of Classical Linear Regression Assumptions
The following are assumptions of the classical linear regression model
(CLRM):
– The regression model is linear, correctly specified, and has an additive
error term
– The error term has a zero population mean-Given X, the expected value of
the error term is zero, or E(ui |X) = 0.
– All explanatory variables are uncorrelated with the error term.
– No Serial Correlation: Observations of the error term are uncorrelated with
each other. cov(ui,uj|X) = 0, i ≠ j.
– No Heteroskedasticity: The error term has a constant variance var(ui|X) =
σ2.
– No explanatory variable/ a perfect linear function of any other explanatory
variable(s).
– The error term is normally distributed, this assumption is optional but
usually is invoked
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 17
1.5 Properties of the Mean, Variance and Standard Error

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 18

Properties of Variance: The mean of sampling distribution to be centered around
the true population, so there is a desirable for the sampling distribution to be as
narrow or precise as possible.
– Centering around “the truth” but with high variability might be of very little
use.
– One way of narrowing the sampling distribution is to increase the sampling
size, hence, also increases the degrees of freedom.

Properties of the Standard Error: Standard error of the estimated coefficient, is

the square root of the estimated variance of the estimated coefficients.
– Hence, it is similarly affected by the sample size such as the larger the
sample size, the more precise the coefficient estimates will be.

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 19

 1.6 The Gauss-Markov Theorem and the Properties of OLS Estimators

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 20

 Unbiased: the OLS estimates coefficients are centered around the true
population values

– The OLS estimators are unbiased if the expected of estimators on

average are equal to their respective true values of parameters.

Efficiency: An estimators are said to be efficient if there is unbiased and

no longer unbiased estimator has a smaller variance.

Minimum variance: no other unbiased estimator has a lower variance for

each estimated coefficient than OLS

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 21

 Consistent: as the sample size gets larger, the variance gets smaller, and each
estimate approaches the true value of the coefficient being estimated

– The OLS estimators are said to consistent if their convergence to their to

true values of parameters as a sample size increases to infinity.

Normally distributed: when the error term is normally distributed, so are the
estimated coefficients.

—This enables various statistical tests requiring normality to be applied.

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 22

 1.7 Method of Ordinary Least Squares

– Ordinary least squares(OLS) is the most common method used to fit a line to
the data for estimation the regression

– The method of OLS does not minimize the sum of the error term, but
minimizes error/residuals sum of squares

 i  i 1 2 2 i 3 3i
u 2
 (Y  B  B X  B X  ....  Bk X ki ) 2

– To obtain values of the regression coefficients, the derivatives are taken with
respect to the regression coefficients and set equal to zero.

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 23

 • For example, a multiple regression model of the two independent
variables is expressed as:

• From the expressed regression model above, the error term is thus
obtained as:

. • Thus, the residuals sum of squared (RSS) is as follows:

ei    Y   0   1 X 1   2 X 2 
  
2

 

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 24

 • The aim of OLS is to obtain the OLS estimators by minimizing the
sum of squared residuals.

• Thus, to obtain the OLS estimators, we need to obtain the first order
partial derivatives of (residuals sum of squares (RSS) and equate them
to zero, i.e. optimization.

  ei 2 
2
  
 0

 
2 
 Y   0   1 X 1   2 X 2
 2
0

  Y   0   1 X 1   2 X 2   0
   
...⇨
  
⇨ Y  n  0   1  X 1   2  X 2  0
....
Thus,
  
n  0   1  X 1   2  X 2   Y ........................................................(1)

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 25

   ei
2

 2  Y   0  1 X 1   2 X 2 . X 1   0
   

1

 

 2.    
 0
  1 0 1 1 1
     
2
Y . X X X X X 2
2   2
2 1

  
 Y .X 1   0 X 1   1  X 1   2  X 1 X 2  0
2

Thus,
  
 0  X 1   1  X 1   2  X 1 X 2   Y . X 1 .........................................(2)
2

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 26

   ei
2

 2  Y   0   1 X 1   2 X 2 . X 2   0
   


2  

2    
 0
 
2 
Y .X 2   0 X 2   1 X 1 X 2   2 X 2  
 2
  
........ Y . X 2   0  X 2   1  X 1 X 2   2  X 2  0
2

  
 0  X 2   1  X 1 X 2   2  X 2   Y . X 2 .......................................(3)
2

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 27

 • The result of optimization is three first-order normal equations which
are now reproduced as follows:
  
n  0   1  X 1   2  X 2  Y
  
 0  X 1   1  X 1   2  X 1 X 2  Y .X 1
2

  
 0  X 2   1  X 1 X 2   2  X 2  Y .X 2 • ,

• We can now represent these equations into matrix form as follows:

 n 
 X 2   0  Y 
 

 X 1

     
 X 1  X1  X 1 X 2 .. 1   Y . X 1 
2

      Y . X 
 X 2 X  X 2   2   2
2
X2
 
1

, October 2024
18 Prepared by Urgaia Rissa(PhD) for MBA 28
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 29
1.8 The Coefficient of Determination in Multiple Regression Analysis

The coefficient of determination in regression analysis measures the goodness

of fit, and is given by the formula:

Explained sum of squares (ESS)

Residuals sum of squares (RSS)

Therefore: TSS=ESS+RSS and divide both sides of the equation by

TSS,
TSS ESS RSS ⇨ 1  ESS  RSS or.. ESS  1  RSS
 
TSS TSS TSS TSS TSS TSS TSS

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 30

Note:R2 quite often takes on values of 0.9 or higher for time series regressions

The limit Cases: R2 = 0 and R2 = 1, graphically we can show as follows

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 31

 Problems with R2 as a goodness of fit measure

There are a number of problem with coefficients of determination:

1. R2 is defined in terms of variation about the mean of y so that if a model

is re-parameterised (rearranged) and the dependent variable changes, R2
will change.
2. R2 never falls if more regressors are added to the regression, e.g. consider:
Regression 1: yt = 1 + 2x2t + 3x3t + ut
Regression 2: y = 1 + 2x2t + 3x3t + 4x4t + ut
Hence, R2 will always be at least as high for regression 2 relative to
regression 1.

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 32

1.9 F- test of goodness of fit for estimated coefficients

Y  1   2 X 2  ...   k X k  u

, at least one is different from

zero
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 33
 – The F test statistic for joint effects of the explanatory variable on dependent
one an be calculated as follows:

ESS  k  1
F  k  1, n  k  
RSS  n  k 
ESS
 k  1 R 2  k  1
 TSS 
n  k    n  k
RSS 1 R 2

TSS
– Note that ESS/TSS is the definition of R2 and RSS/TSS is equal to (1 – R2).
– It can be expressed in terms of R2 by dividing the numerator and denominator
by TSS, the total sum of squares.

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 34

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 35
 H0: 2 3 = 2 or H0: 2 2 = 1

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 36

1.10 The relationship between the t- and F- statistic test distribution
– In the multiple regression model there is a difference between the roles
of the F and t tests.
– The F test tests the joint explanatory power of the variables, while the t
tests test their explanatory power individually.
– In the simple regression model the F test was equivalent to the (two-
sided) t test on the slope coefficient because the ‘group’ consisted of just
one variable.
– The F- distribution can be regarded as the equivalent extension of the t
distribution when there is more than one variable but small sample size.
– Any hypothesis which could be tested with a t-test could have been
tested using F-test, but not the other way round.

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 37

 1.11 Regression Analysis and Analysis of Variance

• Analysis of variance (ANOVA) is a study of total sum of squares, (TSS)

and its components, i.e., explained sum of squares (ESS) and residuals
sum of squared (RSS):
ESS + RSS = TSS
 

  
y  u2
 y 2 2

• By dividing the sum of squares (SS) by their associated degrees of

freedom (df), we get the mean sum of squares (MSS).

• The ANOVA table therefore shows the sum of squares (SS), degrees of
freedom (df), mean sum of squares (MSS) and source of variation.

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 38

ANOVA Table
Source of variation Sum of squares (SS) df Mean sum of squares (MSS)

Due to regression   k-1 

(ESS)
 y or   x
2 2 2
ESS  2  x 2
  MSS reg
df k 1

Due to residuals  n-k 

u 2
RSS  u

2
 MSS res
(RSS)
df n  k

Total (TSS) y 2
n-1
F
MSS reg
MSS res

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 39

 • From the ANOVA table, the F statistic is computed as follows:
Mean sum of squares due to regression ESS k  1
F 
RSS
Mean sum of squares due to residual nk

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 40

 Example, Table 1: Supply of a certain quantity in Addis Ababa City

Year Price(X) Quantity

(000) Demanded(Y)
(000)
2017 10 40
2018 12 50
2019 15 60
2020 20 70

2021 25 80
2022 30 90
2023 40 100

Using this hypothetical data, we can calculate the terms as shown in table 2 below:

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 41

 Table 2: Calculation of Terms of the Linear Equations in Simple Regression

Year ΣXt ΣYt ΣXt2 ΣXtYt

2017 10 40 100 400
2018 12 50 144 600
2019 15 60 225 900
2020 20 70 400 1400
2021 25 80 625 2000
2022 30 90 900 2700
2023 40 100 1600 4000

t=7 152 490 3994 12000

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 42

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 43
 Using the information, we can now obtain the OLS estimators for constant
over 10 observations

 X  550, Y  90,  XY  6,340,  X 2

 38,500,  Y 2  1,054

Solution

1 
n Yi X i   Yi . X i

10  6340  55  90
= 0.1685
n X i   X i  10  38500  550
2 2 2


Y  0.2667  0.1685 X

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 44

Interpretation of the results:
– Holding X constant, the expected or mean value is -0.2667(which a loss).
– An increase in X by one unit, will lead to an increase in Y by 0.1685
units.
Assume that given X= 150, we can now predict Y as follows:

Y  0.2667  0.1685 X

Y  0.2667  0.1685(150)

Y  25

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 45

Example: Given the following data, regress Y on X1 and X2.
Y 6 10 9 14 7 5

X1 1 3 2 -2 3 5
X2 3 -1 4 6 2 4

SOLUTION

Y X1 X2 Y2 X12 X22 YX1 YX2 X1X2

6 1 3 36 1 9 6 18 3
10 3 -1 100 9 1 30 -10 -3
9 2 4 81 4 16 18 36 8
14 -2 6 196 4 36 -28 84 -12
7 3 2 49 9 4 21 14 6
5 5 4 25 25 16 25 20 20
12 18 487 52 82 72 162 22

n=6.
Thus, we can write as follows:-
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 46
 n   
 X1  2    0   Y  6 12 18    51 


 X
   0  
          72 
22..


 X 1 X  X 1 X 2 .. 1   Y . X 1  ⇨ 
2
1 12 52  1   
      Y . X     
  
 X 2 X  X 2   2   2 82    2  162
2
X2 18 22

1
      

According to Cramer’s rule:

51 12 18 6 51 18

72 52.... 22 12 72 22

162 22 82 18 162 82

41,580 
 4,284
0    11.7857 1    1.2143
6 12 18 3,528 3,528
6 12 18
12 52 22 12 52 22
18 22 82
18 22 82

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 47

 6 12 51

12 52 72 Hence, the regression of Y, X1, and

X2 is as follows:
18 22 162

 1,008
0    0.2857
6 12 18 3,528 Y  11.7857  1.2143X 1  0.2857 X 2

12 52 22

18 22 82

Interpretation:

 0  11.7857, the average or expected value of Y is 11.7857, holding X1 and
X2 unchanged.

1 

 2  0.2857 , Holding X2 constant, a unit increase in X3, will lead to a 0.2857
decrease in Y.
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 48
 Finding the OLS estimators using Deviation Method
:
We can also obtain the OLS estimator using the deviation method as follows:
,
 x2
 x1 x2 ..  1    yx1 
  

 1  _  _  _

 2  

    0  Y  1 X 1  2 X 2
 x1 x2  x2    2  yx2 
,
Where :

X1 
_ _
X1
x1  X 1  X 1
n
X2 
_ _
X2
x2  X 2  X 2
n
Y2 
_ _
Y
y Y Y
n

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 49

 The deviation method is now illustrated below:

6 1 3 -2.5 -1 0 1 0 0 2.5 0

10 3 -1 1.5 1 -4 1 16 -4 1.5 -6

9 2 4 0.5 0 1 0 1 0 0 0.5

14 -2 6 5.5 -4 3 16 9 -12 -28 16.5

7 3 2 -1.5 1 -1 1 1 -1 -1.5 1.5

5 5 4 -3.5 3 1 9 1 3 -10.5 -3.5

12 18 28 28 14 51 9

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 50

 Y 
_
Y 51
  8.5
n 6

X1  1   2
_
X 12
Note:
n 6

X 2   2  18 / 6  3
_
X
n

 x2
 x1 x2 .. 1    yx1 
  

 1 28   
 14   1   30
 

 2  

    ..     
 x1 x2  2   2   2 
x  yx 
 14
  
28   2   9 
 

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 51

 Hence, using crammers rule:
28  30
 30  14
  14 9 168

1 
9 28

714
 1.2143 2    0.2857
28  14 588
28  14 688
 14 28
 14 28
   _  _
 0  Y  1 X 1  2 X 2

 0  8.5  1.2143.(2)  0.2857.(3)

 0  11.7857

Thus, the multiple regression models is:

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 52

  
ESS   1  yx1   2  yx2
ESS  1.2143.(30)  0.2857.(9)
ESS  33.8577

Total Sum of Squares (TSS) is obtained by  y 2

y -2.5 1.5 0.5 5.5 -1.5 -3.5

y
2
6.25 2.25 0.25 30.25 2.25 12.25 y 2
 53.5

Thus, TSS =  y  53.5

2 ESS 33.8577
R2    0.6329
TSS 53.5
Interpretation:
Holding all other factors constant, the variables X1 and X2 account for
63.29% of all changes or variations in Y.
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 53
The adjusted R2 is obtained as:

 2 n 1

R  1  1  R 
2

 n  k 

  6  1 
R  1  1  0.6329
2

  6  3 

R 2  1  0.6119

R 2  0.3881
Interpretation:
Holding all other factors constant, X explains or accounts for 38.8% of changes
in Y, when adjusted for degrees of freedom.

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 54

1.12 Tests for the Regression Coefficients ,Statistical Inference using
Hypothesis Testing

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 55

1. Choose a significance level, , (say 5%). This is equivalent to choosing a
(1-)100% confidence interval, 95% confidence interval

2. Use the t-tables to find the appropriate critical value, which will again have
T-K degrees of freedom.

3. The confidence interval is given by

4. If the hypothesised value of  or * lies outside the confidence interval,

then reject the null hypothesis that  =*, otherwise do not reject the null.

Note that the test of significance and confidence interval approaches always
give the same answer.

Under the test of significance approach, we would not reject H0 that  = * if

the test statistic lies within the non-rejection region.
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 56
 • Fore instance, using the regression results below, test the significance of
the coefficients

• Using both the test of significance and confidence interval approaches, test
the hypothesis that  =1 against a two-sided alternative.

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 57

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 58
This is test of significant approach. Therefore, we don’t reject the null hypothesis

• Confidence interval approach:

ˆ  t crit  SE( ˆ )
 0.5091 2.086  0.2561
 (0.0251,1.0433)

• The decision is that do not reject H0 since test stat lies within non-rejection
region.

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 59

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 60
 p-Values

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 61

 ESS / df R 2 /( k  1)
F  
RSS / df (1  R 2 ) /( n  k )

18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 62

Example 1:

Since the test statistic in both cases are greater than the tabulated values, we reject
the null hypothesis at the conventional levels of 1% and 5% significances.
Therefore, we conclude that the model is adequate
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 63
 Example 2: We assume that unrestricted regression model is given as :

yt = 1 + 2x2t + 3x3t + 4x4t + ut.

Then we want to test the restriction that 3+4 = 1.

What is the restricted regression?

Solution:
yt = 1 + 2x2t + 3x3t + 4x4t + ut s.t. 3+4 = 1

We substitute the restriction (3+4=1) into the unrestricted regression

so that it is automatically imposed on the data.
3+4 = 1  4 = 1- 3
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 64
where URSS and RRSS denote unrestricted and restricted residual sums of
regression, m is number of restriction coefficients, T= number of observations and
k= number of regressors in unrestricted regression including a constant or the total
number of parameters to be estimated.
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 65
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 66
18 October 2024 Prepared by Urgaia Rissa(PhD) for MBA 67