Problem 5.19

Refer to S.A.T. data given in exercise 2.16

Suppose you want to predict the male math score  on the basis of the female math score (X) by running the following regression:

Yt = £]1+£]2Xt +ut

exercise 2.16

Download:  0519.dat0519.xls
 

year malesv femalesv totalv malem femalesm totalm
1967
463
468
466
514
467
492
1968
464
466
466
512
470
492
1969
459
466
463
513
470
493
1970
459
461
460
509
465
488
1971
454
457
455
507
466
488
1972
454
452
453
505
461
484
1973
446
443
445
502
460
481
1974
447
442
444
501
459
480
1975
437
431
434
495
449
472
1976
433
430
431
497
446
472
1977
431
427
429
497
445
470
1978
433
425
429
494
444
468
1979
431
423
427
493
443
467
1980
428
420
424
491
443
466
1981
430
418
424
492
443
466
1982
431
421
426
493
443
467
1983
430
420
425
493
445
468
1984
433
420
426
495
449
471
1985
437
425
431
499
452
475
1986
437
426
431
501
451
475
1987
435
425
430
500
453
476
1988
435
422
428
498
455
476
1989
434
421
427
500
454
476
1990
429
419
424
499
455
476

(a) Estimate the preceding model (Eviews)

(b) From the estimated residuals, find out if the normality assumption can be sustained?
     ( Suggested Answer)

(c) Now test the hypothesis that £]2 = 1 , that is, there is a one to one correspondence between
     male and female math scores. (Suggested Answer)

(d) Set up the ANOVA table for this problem. (Suggested Answer)
 

back to mainpage