Economics and Business

Academic Year 2017-2018

Exam: Econometrics 1
6012B0374Y
Date and time of the exam: 27 March 2018 , 13:00-15:00

Duration of the exam: 2 hours

Identification:

You have to identify yourself using your certificate of registration (UvA-identification card) and a valid proof
of identity (passport, ID card) with a good resembling photograph. If you cannot identify yourself, access to
the exam may be denied.

If you are not correctly registered via SIS for the course component, your exam will not be marked and
registered.

Please write your name, student number and serial number on every sheet you hand in.

Warning against fraud/cheating:

Students who are caught at any form of fraud/cheating will be punished.

Make sure that your mobile phone is switched off and put away in your briefcase/bag. This also applies for
other audio equipment, headphones, digital watches (e.g. I-watches) and other electronic devices. Your
briefcase/bag must be closed and placed on the floor beside your table.

Tools allowed:
pencil and pen (not red!), eraser, ruler, non-graphical calculator

Bathroom visits:
It is not allowed to visit the bathroom during the exam.

Specific information on this exam: 2 questions on 2 pages. Maximum 100 points. For each sub-question the
maximum points are indicated. Besides this front page and the questions you also get a page with statistical tables and
two pages with formulas. In total 6 printed pages.
IT IS NOT ALLOWED TO TEAR OFF SHEETS
ANSWER QUESTIONS 1 AND 2 ON TWO DIFFERENT FOLIO SHEETS
Always use α=5% unless indicated otherwise.
Always motivate your answers, give all required steps and indicate used theorems.
 Statistical Tables Econometrics 1
Percentiles of the Student-t Distribution
(Tabulated is t such that P[ T > t ] = P)
dof 0.10 0.05 0.025 0.005
1 3.078 6.314 12.71 63.66
2 1.886 2.920 4.303 9.925
3 1.638 2.353 3.182 5.841
4 1.533 2.132 2.776 4.604
5 1.476 2.015 2.571 4.032
10 1.372 1.812 2.228 3.169
30 1.310 1.697 2.042 2.750
100 1.290 1.660 1.984 2.626
∞ 1.282 1.645 1.960 2.576

Percentiles of the Chi-Squared Distribution

(Tabulated is c such that Prob(χ2dof > c ) = P)
dof 0.10 0.05 0.01
1 2.71 3.84 6.63
2 4.61 5.99 9.21
3 6.25 7.81 11.34
4 7.78 9.49 13.28
5 9.24 11.07 15.09
6 10.64 12.59 16.81
7 12.02 14.07 18.48
8 13.36 15.51 20.09
9 14.68 16.92 21.67
10 15.99 18.31 23.21
15 22.31 25.00 30.58
25 34.38 37.65 44.31
100 118.50 124.34 135.81

Upper 5% points of the F-distribution

m = degrees of freedom for the numerator, n = degrees of freedom for the denominator
m
n 1 2 3 4 5 6 7 8 9 10 15 25 50 100 ∞
1 161.4 199.5 215.7 224.6 230.2 234. 236.8 238.9 240.5 241.9 245.9 249.3 251.8 253. 254.3
2 18.51 19.00 19.16 19.25 19.3 19.33 19.35 19.37 19.38 19.4 19.43 19.46 19.48 19.49 19.5
3 10.13 9.55 9.28 9.12 9.01 8.94 8.89 8.85 8.81 8.79 8.70 8.63 8.58 8.55 8.53
4 7.71 6.94 6.59 6.39 6.26 6.16 6.09 6.04 6.00 5.96 5.86 5.77 5.70 5.66 5.63
5 6.61 5.79 5.41 5.19 5.05 4.95 4.88 4.82 4.77 4.74 4.62 4.52 4.44 4.41 4.36
6 5.99 5.14 4.76 4.53 4.39 4.28 4.21 4.15 4.10 4.06 3.94 3.83 3.75 3.71 3.67
7 5.59 4.74 4.35 4.12 3.97 3.87 3.79 3.73 3.68 3.64 3.51 3.40 3.32 3.27 3.23
8 5.32 4.46 4.07 3.84 3.69 3.58 3.50 3.44 3.39 3.35 3.22 3.11 3.02 2.97 2.93
9 5.12 4.26 3.86 3.63 3.48 3.37 3.29 3.23 3.18 3.14 3.01 2.89 2.80 2.76 2.71
10 4.96 4.10 3.71 3.48 3.33 3.22 3.14 3.07 3.02 2.98 2.85 2.73 2.64 2.59 2.54
15 4.54 3.68 3.29 3.06 2.90 2.79 2.71 2.64 2.59 2.54 2.40 2.28 2.18 2.12 2.07
25 4.24 3.39 2.99 2.76 2.60 2.49 2.40 2.34 2.28 2.24 2.09 1.96 1.84 1.78 1.71
50 4.03 3.18 2.79 2.56 2.40 2.29 2.20 2.13 2.07 2.03 1.87 1.73 1.60 1.52 1.44
100 3.94 3.09 2.70 2.46 2.31 2.19 2.10 2.03 1.97 1.93 1.77 1.62 1.48 1.39 1.28
∞ 3.84 3.00 2.60 2.37 2.21 2.10 2.01 1.94 1.88 1.83 1.67 1.51 1.35 1.24 1.00
         +453+$65+0/4 6/&'3 5*' 45#/&#3& #446.15+0/4

+/'#3 3')3'44+0/ .0&'- #      
 
 

6  -  * 2   
 #   
6 #& 3  3-  *
2   

 & &


6-5+%0--+/'#3+59

& 


#    
6 03 * +/%- %0/45#/5  "1# 
& &
-  2
 
2 
-* 03 *  +/%- %0/45#/5  "1#   ) *
2
1'%+#- .#53+%'4
-    

   
 
    
  '453+%5+0/4   0
   -    
  0
#  0     #  0' 

 ' -  *
    +7'/ 5*' #644#3,07 #446.15+0/4
 

& &-  *
5*'  '45+.#503 # +4 
  0
& &  & &' 
00&/'44 0( +5 & &-  *
 ' -  *
& & 
 
& &-  *
  0
     ' 
6  6 6   6-      -  *
 ' -  *
'453+%5'& 3')3'44+0/ *08 03'%#45 5'45   
6     6    

#  


 
 6 #   





 # 6    
      

3+4%* #6)* *08 3'#, 5'45    

6     6    
6 
6    
      

# 3'&+%5+0/
   6  6

*+426#3'& &+453+$65+0/ 
  

6 #     
       6

' 
( %      8+5* % '   #/&  1& 5*'/ %   %  6  6      

  
( %     8+5* % -   #/&  -  - +4 /0/450%*#45+% 
49..'53+% +&'.105'/5 8+5* 1 31 5*'/ % %   1 6  6       
 6  6   ,2  , -  *


'..# ! #/& +/&'1'/&'/5 #/&   "  '45+.#5034
   $.45   (03 )  *
 !!  ! !  # 
   6
'/53#- -+.+5 5*'03'. #// #-& ( /+(,  - 
 

 &  #  2 &  & 

1"-* *     "1   /+(, - 
 6 -  *
5*'/ 6/&'3 8'#, 3')6-#3+59 %0/&+5+0/4  &+453+$65+0/4
  
   -#         



 
- 
 
#        

3#.;3:4 5*'03'. ( 7    #/& /+(,  5*'/ #    
3  3-  *

  7     
2     


49.1505+% &+453+$65+0/    
   
# 
#   0   0' 



 

 ' -  *

-#     
 & &  -  *

 5'45
'-'%5+0/ %3+5'3+#
6 6  
5         
6    

    
0-&('-&6#/&5 5'45
 
  


     
  
   2 2   -  * -  *
3'64%*#)#/ 5'45
   

  
 
 6 
    & 7     
 6  6  6
-  /  
- -
 

*+5' 5'45 #4 #$07' 8+5*

& & / & & / -
/    /    7  5 5 5 5  03 8+5* %3044 5'3.4
- - - -

 +(  "1   3'64%*0&(3'9 5'45
  
#  
 6 & 5   &    &    -  / 03  5'45
#3)#/ 5'45
 


#   
 6
 
 

 %0/&+5+0/4 )+7'/ ! -  , &  !    -  ,  *

/+(, !  -  #64.#/ 5'45   %0/5#+/4 * %0-6./4 0( 

/+(, !  -  1"-*  *    !!  ! !   
 
/+(, !  !-  1"-*  , &        -  *  03  5'45
Final exam Econometrics 1 (6012B0374Y), 27 March 2018, 13:00-15:00

Question 1 (50 points)

Consider the classical linear regression model

=y Xβ +ε ,
where y: n×1, X: n×k with rank(X)=k and plim(n-1X’X) = Q positive definite, β: k×1 a vector of constant
parameters, and ε : n×1 =
with E (ε | X ) 0,=Var ( ε | X ) σ 2 I n .
ANSWER QUESTIONS 1 AND 2 ON TWO DIFFERENT FOLIO SHEETS
a. (10p) Derive the least squares (OLS) estimator b of β.
b. (15p) Derive a statistic and its asymptotic distribution that can be used to make a confidence
region for Rβ (R:g×k, rank(R) = g) for large n, and also give the 95% confidence region. In your
derivation you may use that √n (b–β)  d
→ N(0,σ2 Q-1) and plim(s2) = σ2 .

In the rest of this question we work conditional on an observed matrix X. We do not make this
conditioning explicit in the notation, but simply assume that X is fixed. In addition we assume that ε is
normally distributed. Note that the model can also be written as
y=
i xi1β1 + ... + xik β k + ε i (=
i 1,..., n ).
ANSWER QUESTIONS 1 AND 2 ON TWO DIFFERENT FOLIO SHEETS
c. (4p) When the variable xi2 is actually redundant (β2 = 0), what are the consequences for
unbiasedness, efficiency and consistency of the OLS-estimator b of β? Explain your answer.
d. (4p) Suppose you suspect that the marginal effect of xi2 on yi depends on the value of xi3 and the
marginal effect of xi3 on yi depends on the value of xi2, which would make the given model
incorrect. Explain how you would test this.
e. (7p) Show that when Var(ε ) = Ω instead of σ2 In , the residuals e and the OLS-estimator b are in
general not independent, while in the special case Ω = σ2 In , e and b are independent.
f. (10p) Specify an efficient estimator of β in matrix notation, when the errors are uncorrelated and
Var(εi ) = σ2vi, where vi is positive, fixed and known. Show that your estimator is efficient.

Question 2 (50 points)

With yearly data of 1970-1999 a model is estimated, using the following variables:
P = consumer price index,
PG = price of gasoline,
Y = log(expenditure on gasoline / PG),
X2 = log(nominal disposable income / P),
X3 = log(PG / P),
X4 = log(price index of public transport / P),
X5 = log(price index of new cars / P),
X6 = log(price index of used cars / P).

A linear model is estimated with results from EViews reported below.

 Dependent Variable: Y
Method: Least Squares
Sample: 1970 1999
Included observations: 30
Variable Coefficient Std. Error t-Statistic Prob.

C 3.992342 0.552295 7.228642 0.0000

X2 0.866051 0.163916 5.283505 0.0000
X3 -0.439604 0.062818 -6.998054 0.0000
X4 0.033513 0.078636 0.426181 0.6738
X5 0.390288 0.247657 1.575924 0.1281
X6 -0.094666 0.059347 -1.595125 0.1238

R-squared 0.989532 Mean dependent var 6.921244

Adjusted R-squared 0.987352 S.D. dependent var 0.203019
S.E. of regression 0.022832 Akaike info criterion -4.544418
Sum squared resid 0.012512 Schwarz criterion -4.264178
Log likelihood 74.16626 F-statistic 453.7607
Durbin-Watson stat 1.208064 Prob(F-statistic) 0.000000

R-squared when regressing each independent variable on all the other independent variables:
Variable: X2 X3 X4 X5 X6
R-squared: 0.986 0.887 0.845 0.984 0.487

ANSWER QUESTIONS 1 AND 2 ON TWO DIFFERENT FOLIO SHEETS

a. (4p) Interpret the estimated coefficient of X2.
b. (4p) What is the p-value when you test whether X6 has a negative effect on Y? What is the p-
value when you test whether X5 has a negative effect on Y?
c. (7p) Which of the estimated coefficients are subject to multicollinearity? What are the
consequences of multicollinearity?
d. (7p) The years 1973-1977 and 1979-1980 were special, because of two oil crises.
Which Chow test would you use to test for a structural difference between those seven
years and the other years. Also, give the formula of the F-statistic of this test in terms
of two residual sums of squares, fill out this formula with numbers as far as possible,
and explain which regression you would need to do in order to get the residual sum of
squares for which you do not yet have a number.
e. (5p) Suppose you consider it likely that the errors are heteroscedastic and auto-
correlated, but you do not know the exact form of the variance matrix of the
vector of errors. How would you go about estimating and testing the
coefficients?
f. (10p) It is suspected that X3=log(PG/P) is itself dependent on Y and thus
endogenous. Show the consequences for the consistency of the OLS-estimator b.
g. (13p) Suppose that X3 = log(PG/P) is itself dependent on Y and thus endogenous,
while X7=log(PO/P), where PO is the price of crude oil, is not dependent on Y and
exogenous.
(i) Specify the estimator of the coefficients that you would use in this case.
(ii) Specify your assumptions regarding X7=log(PO/P).
(iii) What are the properties of the estimator (unbiased, consistent, normal)?