MULTIPLE REGRESSION

1. You are given the following data:

Y 1 3 8

X2 1 2 3

X3 2 1 –3

Obtain the estimated regression equation using ordinary least squares if Y is

regressed on X 2 and X 3 with an intercept term.

2. You are given the following data

Y X2 X3

1 1 2

3 2 1

8 3 –3

Based on these data, estimate the following regressions:

(a) Yi  A1  A2X2i  u i

(b) Yi  C1  C3X3i  u i

(c) Yi  B1  B2 X 2i  B3X3i  u i

1. Is A 2  B2 ? Why or why not?

2. Is C3  B3 ? Why or why not?

What general conclusion can you draw from this exercise?

3. You are given the following data based on 15 observations:

Y  367.693; X 2  402.760; X 3  8.0; y 2

i  66042.269

Bliss Point Studies 1 Amit

9811343938 9891555578
 x 2
2i  84855.096; x 2
3i  280.0; y xi 2i  74778.346

y x i 3i  4250.9; x 2i x 3i  4796.0

where lowercase letters, as usual, denote deviations from sample mean values.

(a) Estimate the three multiple regression coefficients.

(b) Estimate their standard errors.

(c) Obtain R 2 and R 2

(d) Estimate 95% confidence intervals for B2 and B3 .

(e) Test the statistical significance of each estimated regression coefficient

using α  5% (two-tail).

(f) Test at α  5% that all partial slope coefficients are equal to zero. Show
the ANOVA table.

4. From the following date estimate the partial regression coefficients, their
standard errors, and the adjusted and unadjusted R 2 values:

Y  367.693 X 2  402.760 X3  8.0 N = 15

 Y  Y   66042.269  X  X 2   84855.096
2 2
i 2i

 X  X 3   280.000  Y  Y X 2i  X 2   74778.346

2
3i i

 Y  YX
i 3i  X3   4250.900  X 2i  X 2 X 3i  X3   4796.000

5. A three-variable regression gave the following results:

Source of variation Sum of squares d.f Mean sum of

(SS) squares (MSS)

Due to regression (ESS) 65965 –– ––

Due to residual (RSS) –– –– ––

Total (TSS) 66042 14 ––

(a) What is the sample size?

Bliss Point Studies 2 Amit

9811343938 9891555578
 (b) What is the value of the RSS?

(c) What are the d.f. of the ESS and RSS?

(d) What is R2? And R 2 ?

(e) Test the hypothesis that X 2 and X 3 have zero influence on Y. Which test
do you use and why?

(f) From the preceding information, can you tell what is the individual
contribution of X 2 and X 3 toward Y?

6. To explain what determines the price of air conditioners, B.T. Ratchford,

obtained the following regression results based on a sample of 19 air
conditioners:

Ŷi  68.236  0.023X2i  19.729X3i  7.653X4i R 2  0.84

se = (0.005) (8.992) (3.082)

where Y = the price, in dollars

X 2 = the BTU rating of air conditioner

X 3 = the energy efficiency ratio

X 4 = the number of settings

se = standard errors

(a) Interpret the regression results.

(b) Do the results make economic sense?

(c) At α  5% , test the hypothesis that the BTU rating has no effect on the
price of an air conditioner versus that it has a positive effect.

(d) Would you accept the null hypothesis that the three explanatory
variables explain a substantial variation in the prices of air conditioners?
Show clearly all your calculations.

7. Based on the data for 1965-IQ to 1983-IVQ (n = 76), the following regression
results are obtained :

Bliss Point Studies 3 Amit

9811343938 9891555578
 Ŷt  10.96  0.93X 2t  2.09X 3t

t = (–3.33) (249.06) (–3.09) R 2  0.9996

F = 83753.7

where Y = the PCE ($, in billions)

X 2 = the disposable (i.e., after-tax) income ($, in billion)

X 3 = the prime rate (%) charged by banks

(a) What is the marginal propensity to consume (MPC)––the moment of

additional consumption expenditure from an additional dollar’s personal
disposable income?

(b) Is the MPC statistically different from 1? Show the appropriate testing
procedure.

(c) What is the rationale for the inclusion of the prime rate variable in the
model? A priori, would you expect a negative sign for this variable?

(d) Is b3 significantly different from 0.

(e) Test the hypothesis that R2 = 0.

(f) Compute the standard error of each coefficient.

9. The following model gives the factors affecting sleep (S) viz. total work (T),
education (E) and age (A):

S = 3638 - 0.148 T - 11.13 E + 2.2 A

s.e (112) (0.017) (5.88) (1.45)

n = 706 R 2 = 0.113

(i) Are the variables individually significant?

Dropping E and A from the model generates the following equation:
S = 3586 - 0.151 T R 2 = 0.103

s.e (39) (0.017)

(ii) Compare the two models. Which model will you prefer?

Bliss Point Studies 4 Amit

9811343938 9891555578
 (iii) Does the inclusion of E and A greatly affect the estimated tradeoff
between S and T?

10. The following regression was obtained for an annual data of 15 years:

Ŷ  45.5  0.35 X 2  0.18 X 3

(s.e.) (0.095) (0.12) R 2  0.94

(i) Test the significance of the coefficients of the explanatory variables.

(ii) Test the overall significance of regression.

(iii) What is the adjusted coefficient of multiple determination? When would

you prefer this over the unadjusted coefficient of multiple determination?

(iv) Compute the adjusted coefficient of multiple determination for the

regression.

11. You are given the following regression results:

Ŷt  16,899  2978.5X 2t R 2  0.6149

t = (8.5152) (-4.7280)

Ŷt = 9734.2 – 3782.2 X 2t + 2815 X 3t R 2 = 0.7706

t = (3.3705) (-6.6070) (2.9712)

(a) Can you find out the sample size underlying these results?

(b) Test between the restricted and unrestricted regression.

12. Consider the three-variable linear regression model discussed in this chapter.

(a) Suppose you multiply all the X 2 values by 2. What will be the effect of
this rescaling, if any, on the estimated parameters and their standard
errors?

(b) Now instead of a, suppose you multiply all the Y values by 2. What will
be the effect of this, if any, on the estimated parameters and their
standard errors?

13. From annual data for the years 1968-1987, the following regression results
were obtained:

Bliss Point Studies 5 Amit

9811343938 9891555578
 Ŷt  859.92  0.6470X 2t  23.195X 3t R 2  0.9776 (1)

Ŷt  261.09  0.2452X2t R 2  0.9388 (2)

Where Y = U.S. expenditure on imported goods, billions of 1982 dollars, X 2 =

personal disposable income, billions of 1982 dollars, and X 3 = trends variable.
True or false: The standard error of X 3 in (1) is 4.2750. Show your calculations.

14. State and explain the assumptions of the multiple linear regression model.

15. Child mortality rates (cmr) for 20 countries was regressed on female literacy
rate (flr) and per capita GDP (pcg). The following results were obtained

Cmr = 263.6 – 0.0056 pcg i - 2.23 flr i

Se (116) (0.0019) (0.2099)

R 2 = 0.7077

(i) How would you interpret the magnitude and signs of the coefficients of
the pcg and flr

(ii) Are they statistically significant independently and jointly.

(iii) If by adding another explanatory variable R 2 increases to 0.75 will it

justify including the new variable in the model.

Question Bank

1. You are given the following data:

Y 1 3 8

X2 1 2 3

X3 2 1 –3

Obtain the estimated regression equation using ordinary least squares if Y is

regressed on X 2 and X 3 with an intercept term.

2. State and explain the assumptions of the multiple linear regression model.

Bliss Point Studies 6 Amit

9811343938 9891555578
3. Consider the following regression for an imaginary country, say Utopia, for a
period of 15 years. Variables are : IMP = Imports, GNP = Gross National Product
and CPI = Consumer Price Index.

Regression 1 :

IMPt  108.20  0.045 GNP2t  0.931CPI3t

t= (3.45) (1.232) (1.844) R 2  0.9894

(i) Test whether, individually, the partial slope coefficients for GNP and CPI
are statistically significant at the 5% level of significance.

(ii) Test whether GNP and CPI jointly have any statistical significance in
explaining variations in imports. Carry out this test at 5% level of
significance.

(iii) Comment on the results obtained in (i) and (ii) above. Do you suspect any
problem?

(iv) Do you expect that OLS coefficients have retained their BLUE property?
If yes, explain why you would still be worried about their quality.

(v) Using a transformation of variables, real imports are regressed on real

income.

Regression 2 :

IMPt GNPt
 1.39  0.202
CPI t CPI t

t= (5.45) (12.22) R 2  0.9142

Would you say that the problem identified above has now been somewhat
solved?

4. An econometric analyst is estimating the following production function from

annual data on a firm in India : Q  β 0  β1L  β 2 K

Where L = Rupees of Labour, K = Rupees of Capital

The analyst knows that the firm always budgets Rs. 12 lakhs a year for labour
and capital together. The other relevant data are provided :

Bliss Point Studies 7 Amit

9811343938 9891555578
 X 2
2i  14588, X3i2  2725,  Yi2  47921, X2i Yi  7454,

X 3i Yi  4554,  X2iX3i  4796, X 2  5802.25, X3  18, Y  67, N  14

Can you estimate the regression coefficients in this model? Explain your
answer.

5. What is Adjusted coefficient of multiple determination? When would you prefer

this measure over the coefficient of multiple determination?

6. Based on a sample of 38 countries the following regression was obtained:

Ŷi  414.4583  0.0523X1i  50.0476X 2i

s e  266.4583 0.0018 9.9581

t = (1.5538) (28.2742) (–5.0257)

R 2  0.9616 R 2  0.9594 F = 439.22

where

Yi = expenditure on education (billions of rupees)

X1 = GDP (billions of Rupees)

X 2 = Population (billions of people)

(i) Test whether the partial slope coefficients of GDP and population are
individually statistically significant at 5% level of significance.

(ii) Test whether jointly both GDP and population significantly explain
variations in the dependent variable. Use α = 5%.

(iii) Now, if we impose the restriction that slope coefficient on population is

zero. We obtain the following regression :

Ŷi  386.482  0.0732X1i

s e  (268.421) (0.0049)
t = (1.4398) (14.9397)

R 2 = 0.8978 R 2 = 0.8823 F = 436.81

Bliss Point Studies 8 Amit

9811343938 9891555578
 Test whether this restriction is statistically significant at 5% level of
significance.

(iv) In comparing the restricted and unrestricted regression, if dependent

variables do not have the same form, which alternative test do you use?

(v) When is it justifiable to add another explanatory variable in the model?

7. The monthly salary (WAGE, in hundreds of rupees), age (AGE, in year), number
of year of experience (EXP, in years), number of years of education (EDU) were
obtained for 49 persons in a certain office. The estimated regression of WAGE
on the characteristics of a person were obtained as follows (with t statistics in
parenthesis) :

WAGE = 632.244 + 142.510EDU + 43.225 EXP – 1.913 AGE

(1.493) (4.088) (3.022) (–0.22)

(i) The value of adjusted R2, R–2 = 0.277. Using this information, test the
model for overall significance at 1% level of significance.

(ii) Test the coefficient of EDU and EXP for statistical significance at 1% level
and coefficient for AGE at 10% level.

(iii) Can you rationalize the negative sign for AGE? If someone suggests that
AGE be eliminated, will you follow the suggestion?

8. Consider the following model relating the gain in salary due to an MBA degree
to a number of its determinants.

SLRYGAIN t = B1  B2 TUTION t  B3 Z1t  B4 Z2t  B5 Z3t  u t

Where

SLRYGAIN = Post salary MBA minus pre MBA salary, in thousands of dollars.

TUITION = annual tuition costs, in thousands of dollars

Z1 = MBA skills in being analysts, graded by recruiters.

Z2 = MBA skills in being team players, graded by recruiters.

Z3 = Curriculum evaluation by MBA’s

Using data for 25 top business schools, the coefficients were estimated as
follows standard errors in parenthesis.

Bliss Point Studies 9 Amit

9811343938 9891555578
 B̂1 60.899 (2.513)

B̂ 2 0.314 (0.750)

B̂3 –3.948 (2.756)

B̂ 4 –2.016 (2.165)

B̂5 –5.325 (3.773)

(i) Carry out individual two tails tests at 10% level of significance for the
slope coefficients.

(ii) Test the model for overall significance at the 10% level if R 2  0.461 was
obtained for the model.

(iii) Is there a conflict between your conclusions in (i) and (ii)? If yes, can you
suggest a possible explanation?

9. For a regression of variable Y, on two explanatory variables X1 and X 2 ,

illustrate the ANOVA (Analysis of Variance) Table.

Bliss Point Studies 10 Amit

9811343938 9891555578