EViews assignments – 4th week

10.4
a) Re-open the exchange rate returns series and test them for day-of-the-week effects.
b) Re-open the house price changes series and determine whether there is any evidence of
seasonality.

12.5 (slightly changed)

Re-open the ‘fail xls’ spreadsheet for modelling the probability of MSc failure and do the following:
a) Take the country code series and construct separate dummy variables for each country.
Re-run the probit and logit regression above with all of the other variables plus the country dummy
variables. Set up the regression so that the UK becomes the reference point against which the
effect on failure rate in other countries is measured. Is there evidence that any countries have
significantly higher or lower probabilities of failure than the UK, holding all other factors in the
model constant?
b) Suppose that a fellow researcher suggests that there may be a non-linear relationship
between the probability of failure and the age of the student. Estimate a probit model with all of
the same variables as above plus an additional one to test this. Is there indeed any evidence of such
a nonlinear relationship?
Question 10.4
(a) Re-open the exchange rate returns series and test them for day-of-the-week effects
r t =γ 1 D 1t + γ 2 D2t +γ 3 D 3t + γ 4 D 4t + γ 5 D 5t + μt

Where:
r t =¿Daily returns
D1 = 1 on Monday and zero otherwise
D2 = 1 on Tuesday and zero otherwise
D3 = 1 on Wednesday and zero otherwise
D4 = 1 on Thursday and zero otherwise
D5 = 1 on Friday and zero otherwise
D6 = 1 on Saturday and zero otherwise
Sundays will be our reference point.

REUR C D1 D2 D3 D4 D5 D6
Method: Least Squares
Date: 12/04/20 Time: 15:54
Sample (adjusted): 6/07/2009 6/06/2013
Included observations: 1461 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.  

C 0.021463 0.029801 0.720209 0.4715

D1 -0.020113 0.042145 -0.477231 0.6333
D2 0.006249 0.042145 0.148270 0.8822
D3 -0.010046 0.042145 -0.238378 0.8116
D4 -0.025300 0.042145 -0.600307 0.5484
D5 -0.024860 0.042195 -0.589172 0.5558
D6 -0.039536 0.042195 -0.936978 0.3489

R-squared 0.001175    Mean dependent var 0.005255

Adjusted R-squared -0.002947    S.D. dependent var 0.430192
S.E. of regression 0.430825    Akaike info criterion 1.158552
Sum squared resid 269.8777    Schwarz criterion 1.183883
Log likelihood -839.3222    Hannan-Quinn criter. 1.168001
F-statistic 0.285066    Durbin-Watson stat 1.397296
Prob(F-statistic) 0.944211
RGBP C D1 D2 D3 D4 D5 D6

Dependent Variable: RGBP

Method: Least Squares
Date: 12/04/20 Time: 15:56
Sample (adjusted): 6/07/2009 6/06/2013
Included observations: 1461 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.  

C 0.012823 0.025324 0.506368 0.6127

D1 -0.009824 0.035814 -0.274301 0.7839
D2 0.013037 0.035814 0.364029 0.7159
D3 -0.009607 0.035814 -0.268246 0.7885
D4 -0.027187 0.035814 -0.759119 0.4479
D5 -0.024811 0.035857 -0.691938 0.4891
D6 -0.009400 0.035857 -0.262149 0.7932

R-squared 0.001226    Mean dependent var 0.003149

Adjusted R-squared -0.002896    S.D. dependent var 0.365580
S.E. of regression 0.366109    Akaike info criterion 0.833009
Sum squared resid 194.8882    Schwarz criterion 0.858340
Log likelihood -601.5131    Hannan-Quinn criter. 0.842458
F-statistic 0.297362    Durbin-Watson stat 1.454486
Prob(F-statistic) 0.938329

RJPY C D1 D2 D3 D4 D5 D6
Dependent Variable: RJPY
Method: Least Squares
Date: 12/04/20 Time: 15:57
Sample (adjusted): 6/07/2009 6/06/2013
Included observations: 1461 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.  

C 0.044940 0.029310 1.533262 0.1254

D1 -0.044492 0.041451 -1.073379 0.2833
D2 -0.093476 0.041451 -2.255114 0.0243
D3 -0.148062 0.041451 -3.572006 0.0004
D4 -0.004379 0.041451 -0.105650 0.9159
D5 -0.024156 0.041500 -0.582072 0.5606
D6 0.012060 0.041500 0.290602 0.7714

R-squared 0.015991    Mean dependent var 0.001674

Adjusted R-squared 0.011930    S.D. dependent var 0.426281
S.E. of regression 0.423731    Akaike info criterion 1.125343
Sum squared resid 261.0624    Schwarz criterion 1.150673
Log likelihood -815.0628    Hannan-Quinn criter. 1.134792
F-statistic 3.938065    Durbin-Watson stat 1.497978
Prob(F-statistic) 0.000653
Summary of the result for testing for day-of-the-week effects
EUR GBP JPY
-0.02146 -0.00982 -0.04449
Monday (0.47723) (0.27430) (1.07338)
0.00625 0.01304 -0.09348
Tuesday (0.14827) 0.36403 (2.25511)*
-0.01005 -0.00961 -0.14806
Wednesday (0.23838) (0.26825) (3.57201)**
-0.02530 -0.02719 -0.00438
Thursday (0.60031) (0.75912) (0.10565)
-0.02486 -0.02481 -0.02416
Friday (0.58917) (0.69194) (0.58207)
-0.03954 -0.00940 0.01206
Saturday (0.93698) (0.26215) 0.29060
Note: Coefficients are given in each cell followed by t-ratios in parenthesis * and ** denote significance at
the 5% and 1% levels respectively. Source: Student’s compilation from e-view output.

Explanation
From the above result, day of the week event is evident for Tuesday and Wednesday only
for JPY as these two days are significance at 5% and 1% respective. On average and
holding everything else equal, the percentage on return significantly decreased on Tuesday
by 0.0935% and by 0.1481% on Wednesday compared to Sunday (reference point).
(b) Re-open the house price changes series and determine whether there is any evidence
of seasonality.

D1 = 1 on January and zero otherwise

D2 = 1 on February and zero otherwise
D3 = 1 on March and zero otherwise
D4 = 1 on April and zero otherwise
D5 = 1 on May and zero otherwise
D6 = 1 on June and zero otherwise
D7 = 1 on July and zero otherwise
D8 = 1 on August and zero otherwise
D9 = 1 on September and zero otherwise
D10 = 1 on October and zero otherwise
D11 = 1 on November and zero otherwise
December will be our reference point.

Dependent Variable: DHP

Method: Least Squares
Date: 12/04/20 Time: 16:19
Sample (adjusted): 1991M02 2013M05
Included observations: 268 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.  

C -0.016242 0.232471 -0.069868 0.9444

D1 -0.318734 0.328764 -0.969493 0.3332
D2 0.570603 0.325171 1.754782 0.0805
D3 1.394375 0.325171 4.288135 0.0000
D4 1.236132 0.325171 3.801488 0.0002
D5 0.847643 0.325171 2.606766 0.0097
D6 1.033996 0.328764 3.145104 0.0019
D7 0.423311 0.328764 1.287586 0.1991
D8 -0.217975 0.328764 -0.663014 0.5079
D9 0.098776 0.328764 0.300448 0.7641
D10 0.117231 0.328764 0.356581 0.7217
D11 0.164045 0.328764 0.498975 0.6182

R-squared 0.209024    Mean dependent var 0.437995

Adjusted R-squared 0.175037    S.D. dependent var 1.200502
S.E. of regression 1.090386    Akaike info criterion 3.054683
Sum squared resid 304.3689    Schwarz criterion 3.215473
Log likelihood -397.3275    Hannan-Quinn criter. 3.119264
F-statistic 6.150082    Durbin-Watson stat 1.427902
Prob(F-statistic) 0.000000

Explanation
From the estimated result above, it is evidence that there is seasonality from March to June.
There is significant increase in house price in March by 1.3943% compared to December.
The house price increased by 1.236% in April compared to December price. 0.8476%
house price increase was recorded in May relative to December price and 1.03399%
increase in June house price as compared to December price.