Assignment of Econometrics Theory and Application

1. Discuss the main reasons behind the inclusion of the error term in the econometric model
2. Consider the model: Y i=β +U i , Where β is a fixed parameter and U i is the
disturbance terms and satisfies all assumptions of a linear regression model. (**)
a. Find the least squares estimate of β .
b. Show that the value of R2 in such a regression is zero.
3. The following table gives aggregate consumption expenditures Y and aggregate
disposable income X of a house hold.
Y 102 106 108 110 122 124 128 130 142 148 150 154
X 114 118 126 130 136 140 148 156 160 164 170 178
a. Fit a linear regression model of Y on X that will help a house hold to predict
consumption expenditure and interpret the coefficients
b. Estimate the constant variance, σ 2 of the error term
c. Estimate Var ( ^β ) , Var ( ^β ) and Cov ( ^β , ^β )
0 1 0 1

d. Test the overall model fit

e. Calculate the R2 and interpret
f. Test whether disposable income has a significant effect on consumption
expenditure at 5% level of significance
g. Construct a 95% confidence interval for true value for β
4. The following table gives the real per capita income in thousands of U.S. dollars Y
with the percentage of the labor force in agriculture X1 and the average years of
schooling of the population over 25 years of age X2 for 15 developed countries in
1981. (a) Find the least-squares regression equation of Y on XI and X2. (b) Interpret
the results of part a. (**)

Y 6 8 8 7 7 12 9 8 9 10 10 11 9 10 11
X1 9 10 8 7 10 4 5 5 6 8 7 4 9 5 8
1
X2 8 13 11 10 12 16 10 0 12 14 12 16 14 10 12
Given

[ ]
−1
3.3912 −0.1541 − 0.1872
( X ' X ) = − 0.1541 0.01720.0028
− 0.1872 0.0028 0.0139
a. Find the least squares regression equation of Y on X1 and X2.
 b. Interpret the results of part a.
c. Test at the 5% level of significance for each of the individual parameters
d. Construct the 95% confidence interval for each of the parameters
e. State the null and alternative hypotheses in testing the overall significance of the
regression.
f. Test the overall significance of the regression model at 5% level of significance
g. Calculate R2 and interpret

5. Consider the following data on per capita food consumption (Y), price of food (X1) and per
capita income (X2) for the years 1950-1964 in the United States. Retail price of food and per
capita disposable income are deflated by the Consumer Price Index. (Using computer
software)

Year Y X1 X2 Year Y X1 X2
1950 88.9 91.7 57.7 1958 85.4 88.1 52.1
1951 88.9 92.0 59.3 1959 88.5 88.0 58.0
1952 89.1 93.1 62.0 1960 88.4 88.4 59.8
1953 88.7 90.9 56.3 1961 88.6 83.5 55.9
1954 88.0 82.3 52.7 1962 91.7 82.4 60.3
1955 85.9 76.3 44.4 1963 93.3 83.0 64.1
1956 86.0 78.3 43.8 1964 95.1 86.2 73.7
1957 87.1 84.3 47.8

a. Fit a linear regression model of Y on X1 and X2

b. Test the overall significance of the model at 5% level of significance
c. Obtain the adjusted coefficient of multiple determination R2 and interpret the
result
d. Does food price significantly affect per capita food consumption at 5% level of
significance level?
e. Does disposable income significantly affect per capita food consumption?
f. Interpret the coefficients
g. Diagnose the model

6. With the help of STATA, using the given sample data in excel:-
A. Fit an OLS regression for the “determinants of crop productivity in Dire Dawa
administration” and interpret the results (only for significant variable(s).
 B. Check for the violation of assumptions of classical linear regression?
I. Check for heteroskedasticity using both Breusch-Pagan / Cook-Weisberg test
and White’s test and report the results?
II. Check for Multicollinearity using variance inflation factor (VIF) and report
the results?
III. Check for omitted variables using Ramsey RESET test and report the results?
7. Discus on the following questions with your groups:
A. What is Multicollinearity? Which assumption of the CLRM is violated?
B. Explain the consequences of MC in simple OLS estimators
C. Explain what the VIF is and its use
D. Show how to detect possible MC in a regression model
E. What happen to the estimate of β when there is perfect collinearity among the X’s?
8. Consider the following outputs for regressing consumption expenditure (Y) on income:
Dependent variable: Y
Number of observation: 30
Variable Coef Std.Error T P-value
Constant 9.290 5.231 1.776 0.087
X 0.638 0.029 22.287 0.000
R-squared 0.947 F=496.718 (P-value=0.000)
Dependent variable: Squared residual
Variable Coef Std.Error T P-value
Constant -12.296.0 191.773 0.064 0.949
X 0.197 2.369 0.083 0.934
SquaredX 0.002 0.007 0.254 0.802
R-squared 0.178
a. Fit the regression model and interpret
b. Test for the existence of heteroscedasticity using the White test at 5% level of
significance
c. Based on the following information, test for the existence of heteroscedasticity using the
Goldfeld–Quandt test at 5% level of significance. First the data ranked on ascending
order by X (income) values and then dropping the middle 4 observations, the OLS
 regressions based on the first 13 and the last 13 observations and their associated residual
sums of squares as shown below (standard errors in the parentheses).
Regression based on the first 13 observations:
Y^ i=3.4094 +0.6968 X i
( 8.7049 )( 0.0744 ) 2
R =0.8887 RSS1=377.17 df =11

Regression based on the last 13 observations:

Y^ i=− 28.0272+0.7941 X i
( 30.5421 ) ( 0.1319 ) 2
R =0.7681 RSS2=1536.8 df =11

9. Consider the following computer output on consumption expenditure (Y) and income (X) for
20 households:
The estimated regression equation of Y on X is:
Expenditure = - 0.574 + 1.10 Income

Predictor Coef SE Coef T P-value

Constant -0.5737 0.7958 -0.72 0.480

Income 1.09632 0.03085 35.53 0.000

R2 = 0.986, F = 1262.64 (p-value = 0.000), Durbin-Watson statistic = 2.55

At 5% level of significance, the Durbin-Watson critical values for 20 observations and one
explanatory variable are dL = 1.20 and dU =1.41.
a. Test for the existence of autocorrelation
b. Suggest some of the methods to estimate ρ if it is not known
10. With the help of STATA, using the given sample data in excel:-
C. Fit an OLS regression for the “determinants of crop productivity in Dire Dawa
administration” and interpret the results (only for significant variable(s).
D. Check for the violation of assumptions of classical linear regression?
IV. Check for heteroskedasticity using both Breusch-Pagan / Cook-Weisberg test
and White’s test and report the results?
 V. Check for Multicollinearity using variance inflation factor (VIF) and report
the results?
VI. Check for omitted variables using Ramsey RESET test and report the results?
VII. Check for normalty using the Jarque Bera test (Skewness/Kurtosis tests) and
report the results?
VIII. Check for autocorrelation using both Durbin-Watson d-statistic and Breusch
Godfrey LM test and report the results?