Example 1:

You are using an econometric model to study the dependence of the annual salaries of
CEOs (Chief Executive Officers) of major private companies on some variables. The
sample data consist of observations for 60 private firms which include the following
variables:
SALi : the annual salary of the CEO of firm i, measured in thousands of dollars;
ARi : the annual total sales revenues of firm i, measured in millions of dollars;
MVi : the market value of firm i, measured in millions of dollars;
EMi : the number of years the CEO has been employed with firm i;
AGEi: the age of the CEO of firm i, in years.
The regression model you propose is:

in which (lnXi) denotes the natural logarithm of X i. EM2 and AGE2 are the squares of
corresponding variables, ui is stochastic disturbance.
Using the data, you estimate the following regression models (estimated standard errors
in parentheses below the coefficient estimates):
(1) lnSALi (hat)=5.572 + 0.182lnARi + 0.102lnMVi + 0.046EMi - 0.00122EMi2 –
0.042AGEi + 0.00033AGEi2
se (0.0412) (0.0493) (0.0142) (0.000476)
(0.0412) (0.00036)
RSS= 42.060; TSS= 64.646
(2) lnSALi (hat)= 4.369 + 0.1646lnARi + 0.1085lnMVi + 0.04512EMi - 0.00121EMi2
RSS= 42.474; TSS= 64.646
1. In the model (1) above, interpret the meaning of each estimated coefficients
Does each independent variable AR or MV affect the salaries of CEOs?
2. In the model (1), by how much the model can explain for the variation of salaries
of CEOs?
Is it correct to say that all independent variables of the model (1) simultaneously
do not explain for the variation of the salaries of CEOs?
3. In the model (1), test the hypothesis that coefficients of AR and MV are equal
given that:
4. State the coefficient restrictions that are imposed on regression equation (1) in
estimating model (2) above? Conduct a test of these coefficient restrictions and
state the meaning of this test? Based on the outcome of the test, would you
choose equation (2) or equation (1)?
5. What are the implications of introducing the squared terms of EM and AGE in the
model (1)? Present the procedure to use F-test to test the hypothesis that we can
drop out two squared terms EM2 and AGE2 from model (1) (use the form of
population regression model).

Example 2:
You want to study the dependence of beer expenditures of employees in a company on
their incomes, ages and sexes. You have collected a random sample of observations on
40 office employees, 20 of whom are females and 20 of whom are males. Here is the
description of variables in the data set:
BEi : the annual beer expenditures of employee i, measured in dollars per year;
INCi : the annual income of employee i, in thousands of dollars per year;
AGEi : the age of employee i, in years;
SEXi : the dummy variable, SEXi = 1 if employee i is female and SEXi = 0 if
employee i is male.
You propose the following model (model (1)):

Using OLS method in EVIEWS, you obtain the following results:

Result (1)
Dependent variable: BE
Included observations: 40

Variable Coefficient Std. Error t-Statistic Prob.

C 489.8631 73.85524 6.632747 0.0000

INC 0.002893 0.000775 3.734180 0.0007

AGE -10.07924 2.229676 -4.520493 0.0001

SEX -265.8574 113.3658 -2.345129 0.0250

SEX*INC -0.001029 0.000971 -1.059491 0.2968

SEX*AGE 4.231494 3.648383 1.159827 0.2542

R-squared 0.6470

Result (2)
BEi = 459.21+ 0.0023 INCi - 8.42 AGEi -169.87 SEXi R2=0.6294
Result (3)
BEi = 342.88+ 0.00238 INCi - 7.575 AGEi R2= 0.3292
1. Write down the sample regression model of model (1) based on the result (1)?
Write down the population regression model and sample regression model for male and
female employees and explain the meaning of the estimated regression coefficients?
2. Using result (1), for male employees, how the expenditures for beer change if
their income increases 1000USD/year? Answer the same question for female
employees given that:
3. In the model (1), state the null and the alternative hypothesis if you want to test
that the models for the expenditures of beer for male and female are not different
in slope coefficients of both INC and AGE. In other words, you want to conduct
the joint test of hypothesis of equal slope coefficients of male and female for INC
and equal slope coefficients of male and female for AGE. Perform this test using
appropriate information given above.
4. Using the results above to test the hypothesis that the variable SEX does not
affect the annual expenditures for beer.
5. Given that d-DW statistic is 1.92. Using this value to test the problem that can be
existed in the model.
Example 3:
In order to explain the US defense budget, you are using the data from 1962 to 1981 with the
following variables (all measured in billions USD) and estimate the corresponding model (Model
1):(Use α=0.05 for references)
Yt: Defense budget outlay for year t
X2t: GNP for year t
X3t: US military sales in year t
X4t: Aerospace industry sales in year t
D1t: Dummy variable presenting the military conflict involving more than 100,000 troops;
D1t=1 if more than 100,000 troops are involved and equal to 0 if fewer than 100,000
troops are involved.

Dependent Variable: Y Sample: 1962 1981

Method: Least Squares Included observations: 20
Variable Coefficient Std. Error t-Statistic Prob.
C 21.40251 1.496947 14.29744 0.0000
D1 -48.21987 6.871544 -7.017328 0.0000
X2 0.013879 0.003207 4.328062 0.0008
X3 0.073146 0.203805 0.358902 0.7254
X4 1.389753 0.130197 10.67423 0.0000
X4*D1 1.540792 0.325005 4.740818 0.0004
X2*D1 0.022406 0.005781 3.876038 0.0019
R-squared 0.996366 Mean dependent var 83.86000
Adjusted R-squared 0.994688 S.D. dependent var 28.97771
S.E. of regression 2.111972 Akaike info criterion 4.602338
Sum squared resid 57.98554 Schwarz criterion 4.950845
Log likelihood -39.02338 F-statistic 593.9815
Durbin-Watson stat 2.233771 Prob(F-statistic) 0.000000

Y = B0 + B1*D + B2*X2 + B3*X3 + B4*X4 + B5*(D*X4) + B6*(D*X2) + u

Y^ = 21.4 – 48.22*D + 0.014*X2 + …

21.4 = ko phải giải thích

1.
D=1: Y^(more than 100k troops) =
D=0: y^(less than 100k troops) =
B1^ = -48.22: defense budget is $48.2 billions lower for the years with more than 100k
troops call compare to that of years with less than 100k troops call
B2: Rate of change of Y by X2 for the years fewer than 100k troops (other variable fixed)
B2 + B6: Rate of change of Y by X2 for the years more than 100k troops (other variable
fixed)
B6: Difference in the rate of change of Y by X between two type of years
B3: Rate of change of Y by X3 when other variables unchanged
 R^2 = 0.996366, 99.63% of varation of defense budget that can be explained by the model
2.
Test for beta j = 0
Ho: beta_j = 0; H1: beta_j khác 0;
Using p-value, compare to alpha = 0.05
P-value < 0.05 -> reject Ho -> the variable is significant
p-value > 0.05 -> the variable is not significant
look at the result

F-test of overall significant:

Ho: model is not significant; H1: model is signi
F test has the p-value = … < 0.05 -> reject Ho ->

3. Durbin-Watson
Breusch Pagan for d-W test

4. X2, k = số tham số
CI 95% for B6

5.
t-test for Test Beta4 + Beta2 = 1.4
CI Beta2 + Beta4
T = ((beta^2 + beta^4) -1.4/ se(beta^2 +beta^4)) (beta^ = beta head, not power)
se(beta^2+beta^4) = sqrt (var(beta^2+beta^4))
var(beta^2+beta^4)= var (beta^2) + var(beta^4) +2cov(beta^2, beta^4)
reject Ho if t > t(alpha/2, n-7) or t < -t
CI: beta ^2 + beta^4
6. F test of dropping variable
Ho: beta1 = beta5 = beta6 = 0
F using R2

7. D1 = 1 if more than, D1 = 0 if fewer than

D2 = 1 fewer than, D2 = 0 more than

D1 + D2 = 1: perfect multicollinearity: cannot estimate the model using OLS method

Imperfect multicollinearity:

8.
1. Explain the meaning of each estimated coefficient and R2 in the above model.
2. Test for significance of each independent variable and test for overall significance of
the model.
3. Conduct the test of autocorrelation in the model using the information above. State
clearly the conditions to apply this test? If those conditions are not met, name other
tests you can use instead.
 4. When GNP increases by 1 bil USD (other variables unchanged), what is the
confidence interval of the difference in the changing levels of defense budget
between the cases of there are more than 100,000 or fewer than 100,000 troops
involved in the military conflict?
5. For the case when there are fewer than 100,000 troops involving in the conflict (this
condition indicates that we are concerning on the coefficients of X2 and X4 only), if
we simultaneously increase X2 and X4 by 1 billions USD, test the proposition that the
defense budget will increase 1.4 billions USD. What is the confidence interval for the
increase in the level of defense budget in this case? (The covariance between two
estimated coefficients of 2 variables X2 and X4 is -0.00036).
6. Do you think that the military budget does not depend on the number of troops
involving in the conflict given that if you regress Y on X2, X3 and X4 (with intercept),
you get R2=0.971 and RSS=461.28?
7. What would happen if the model included dummies for both cases of there are
more than 100,000 and fewer than 100,000 troops involved in the military conflict?
What is this problem called? What are the consequences of this problem?
8. Given the information below, test for all possible problems in the model 1 above. In each
test specify clearly type of problem, name of test, null and alternative hypothesis and
conclusion about the problem.
Result (1)

7
Series: Residuals
6 Sample 1962 1981
Observations 20
5
Mean -1.13E-14
4 Median 0.140813
Maximum 2.855637
Minimum -3.215228
3
Std. Dev. 1.746960
Skewness -0.218699
2
Kurtosis 2.211793
1 Jarque-Bera 0.677155
Probability 0.712783
0
-4 -3 -2 -1 0 1 2 3

Result (2)
White Heteroskedasticity Test: (No cross term)
F-statistic 2.379399 Probability 0.114212
Obs*R-squared 15.31799 Probability 0.168397
Result (3)
Breusch-Godfrey Serial Correlation LM Test: AR(2)
F-statistic 1.950537 Probability 0.188349
Obs*R-squared 5.235963 Probability 0.072950
Res ult (4)

Ramsey RESET Test:

F-statistic 2.110154 Probability 0.119102
Log likelihood ratio 7.432899 Probability 0.059308
Diagnotic test:
Ho: no problem
H1: problem exists
p-value < alpha:
each problem: 4 aspects: definition, consequences, how to detect,
how to correct (theoretically)

logistic regression model:

explain the meaning of coefficient (use the logit)
find the probability p for a specific observation (plug in the logistic
model)
Find the change of probability when 1 variable increase 1 unit:
=beta*p*(1-p)
Test for significant of coefficient

Ko hỏi về panel data