City Sales Radio Newspaper Radio*News

1 973 0 40 0
2 1119 0 40 0 Yi=117,62+11,46X1+18,61X2
3 875 25 25 625 b1: sales will decrease, on average, by 11.46
4 625 25 25 625 thousands $ for each thousand $ increase in radio
5 910 30 30 900 advertising.
6 971 30 30 900 b2: sales will decrease, on average, by 18.61
7 931 35 35 1225 thousands $ for each thousand $ increase in
8 1177 35 35 1225 newspaper advertising.

9 882 40 25 1000
10 982 40 25 1000
b0: When radio & newspaper advertising are null,
11 1628 45 45 2025 sales would be 117.62 thousands $
12 1577 45 45 2025
13 1044 50 0 0
14 914 50 0 0 SUMMARY OUTPUT
15 1329 55 25 1375
16 133 55 25 1375 Regression Statistics
17 1405 60 30 1800 Multiple R 0.7226484082
18 1436 60 30 1800 R Square 0.5222207218
19 1521 65 35 2275 Adjusted R Square 0.4719281662
20 1741 65 35 2275 Standard Error 301.89556603
21 1866 70 40 2800 Observations 22
22 1717 70 40 2800
ANOVA
df
Regression 2
Residual 19
Total 21

Coefficients
Intercept 117.6186965
X radio 11.460362766
X newspaper 18.607642498

RESIDUAL OUTPUT
RESIDUAL OUTPUT
Observation Predicted Y
1 861.92439643
2 861.92439643
3 869.31882812
4 869.31882812
5 1019.6588544
6 1019.6588544
7 1169.9988808
 8 1169.9988808
9 1041.2242696
10 1041.2242696
11 1470.6789334
12 1470.6789334
13 690.63683482
14 690.63683482
15 1213.1297111
16 1213.1297111
17 1363.4697374
18 1363.4697374
19 1513.8097638
20 1513.8097638
21 1664.1497901
22 1664.1497901
 SUMMARY OUTPUT

t critical= 1.729 Regression Statistics

se, on average, by 11.46 5.68107064603673 Multiple R 0.4975269058
housand $ increase in radio 17.2396548869207 R Square 0.247533022
ertising. 5.68<=b1<=17.24 Adjusted R Squ 0.2099096731
se, on average, by 18.61 Standard Error 369.27389311
h thousand $ increase in t STATs > 1.729 Observations 22
er advertising. Reject H0
Significant variables ANOVA
Significant variables df
spaper advertising are null,
117.62 thousands $ Regression 1
Ex1 Residual 20
Ex2 Total 21
Ex3 a
Ex 3 b Coefficients
Ex 4 Intercept 719.79696133
X radio 10.442596685
r(y2|1)^2 0.116470570538651
r(y1|2)^2 0.103501985980544

11.65% of sales are explained by radio advertising and

10.35% are explained by newspaper advertising

SS MS F Significance F
1892752.640697 946376.32035 10.3836584882012 0.00089672535
1731677.722939 91140.932786
3624430.363636

Standard Error t Stat P-value Lower 95% Upper 95% Lower 95,0%
240.8221600818 0.4884047899 0.630852388171462 -386.42787739 621.66527039 -386.42787739
3.342563401065 3.4286149255 0.0028160752773372 4.46429716466 18.456428368 4.4642971647
5.630002681441 3.3050859033 0.0037224529208283 6.82391145961 30.391373537 6.8239114596

durbin watson 7.9483888155

Residuals SQ.Diff SQ residual Negative autocorrelation
111.07560357 - 12337.7897084447
257.07560357 21316 66087.865950891 Residuals
5.681171881755 63199.160284 32.2757139500444 600
-244.318828118 62500 59691.6897730725 400
200
-109.658854442 18133.308511 12025.0643575036
0
-48.6588544419 3721 2367.68411559527 -200 0 5 10 15 20 25
-238.998880765 36229.325621 57120.4650071611 -400
-600
-800
-1000
-1200
 400
200
0
-200 0 5 10 15 20 25
-400
7.001119234501 60516 49.0156705357023 -600
-800
-159.224269615 27630.879898 25352.3680345657 -1000
-59.2242696154 10000 3507.51411148059 -1200
157.3210665872 46891.882631 24749.9179921491
106.3210665872 2601 11304.1692002499 The assumptions are valid
353.363165176 61029.798475 124865.52650317
223.363165176 16900 49891.1035574211
115.8702888874 11554.718453 13425.9238468483
-1080.12971111 1430416 1166680.1928282
41.53026256377 1258121.0965 1724.76270861545
72.53026256377 961 5260.63898756903
7.19023624014 4269.31904 51.6994971890283
227.1902362401 48400 51615.4034428508
201.8502099165 642.11693408 40743.5072433405
52.85020991651 22201 2793.14468821954
 SS MS F Significance F
897166.20093 897166.2 6.579239 0.018471
2727264.1627 136363.2
3624430.3636

Standard Error t Stat P-value Lower 95%Upper 95% Lower 95,0% Upper 95,0%
192.62490388 3.73678 0.001301 317.9885 1121.605 317.988452798845 1121.605469853
4.0711833151 2.565003 0.018471 1.950257 18.93494 1.95025710257219 18.93493626759

SUMMARY OUTPUT

Regression Statistics
Multiple R 0.734542568598
R Square 0.539552785083
Adjusted R Square 0.462811582597
Standard Error 304.4903635218
Observations 22
Upper 95,0%
621.66527039 ANOVA
18.456428368 df
30.391373537 Regression 3
Residual 18
Total 21

Coefficients
Intercept -685.102156678
uals Radio 28.39211689173
Newspaper 40.83216039719
Radio * newspaper -0.46679106818

15 20 25
15 20 25
 SUMMARY OUTPUT

Regression Statistics
Multiple R 0.47604218306
R Square 0.22661616005
Adjusted R Square 0.18794696805
Standard Error 374.371197627
Observations 22

ANOVA
df SS MS F
Regression 1 821354.4913793 821354.491379 5.860380016884
Residual 20 2803075.872257 140153.793613
Total 21 3624430.363636

Coefficients Standard Error t Stat P-value

Intercept 665.847962382 223.3083784741 2.98174196119 0.007372544735
X Newspaper 16.8293103448 6.951898614714 2.42082217787 0.025121435271

p-value<alpha
the interaction term makes a statistically significant contribution
to the model. There is evidence that the interaction between radio and
newspaper advertising affects sales.

SS MS F Significance F 53.9% of the variation in prices

1955571.497039 651857.165679653 7.03080962512 0.002507174337 radio, newspaper, and the ra
1668858.866597 92714.3814776336 52.2% of the variation in prices
3624430.363636 radio and newspaper. The firs

Standard Error t Stat P-value Lower 95% Upper 95% Lower 95,0%
1004.991999139 -0.681699115281848 0.50410683936 -2796.51199793 1426.30768458 -2796.51199793
20.84425852484 1.36210730920956 0.18996771152 -15.4000452556 72.184279039 -15.4000452556
27.59047825137 1.47993666601844 0.15618031186 -17.1332834619 98.7976042563 -17.1332834619
0.567088809533 -0.823135742286051 0.42120292593 -1.6582004469 0.72461831055 -1.6582004469
 Significance F
0.02512143527

Lower 95% Upper 95%Lower 95,0%

Upper 95,0%
200.034847417 1131.661 200.0348 1131.661
2.32790394544 31.33072 2.327904 31.33072

9% of the variation in prices is explained by the

adio, newspaper, and the radio*newspaper.
2% of the variation in prices is explained by the
adio and newspaper. The first model is better.

Upper 95,0%
1426.30768458
72.184279039
98.7976042563
0.72461831055
Price Rooms Neighbourhood SUMMARY OUTPUT
305.70 6 0 yi=246.49+8.89X1+15.81X2
307.5 8 0 Regression Statistics
340.2 9 0 Multiple R 0.9151407086 b1: the price will decrease, on average,
for each increase in rooms.
346.5 12 0 R Square 0.8374825166
308.2 8 0 Adjusted R Square 0.8183628126
b2: the price will be, on average, 15.81
338.8 9 0 Standard Error 8.3513397206 for a west side neighbourhood than an
334.1 11 0 Observations 20
312.2 8 0
327.8 9 0 ANOVA
335.4 9 0 df SS
319.4 7 1 Regression 2 6109.939122807
383.8 13 1 Residual 17 1185.662877193
339.9 10 1 Total 19 7295.602
348.7 10 1
346.1 9 1 Coefficients Standard Error
327.2 8 1 Intercept 246.48725146 10.7081455654
332.9 8 1 X rooms 8.8935672515 1.165997093816
345.8 9 1 X neighbourhood 15.813216374 3.780061154465
363.3 11 1
351.9 9 1

RESIDUAL OUTPUT
q a&b 20
qc Observation Predicted Y Residuals
15
qd 1 299.84865497 5.85134502924
qe 2 317.63578947 -10.13578947368 10
qf 3 326.52935673 13.67064327485
q g&h 4 353.21005848 -6.710058479532 5
qi 5 317.63578947 -9.435789473684 0
qj 6 326.52935673 12.27064327485 0 5

7 344.31649123 -10.21649122807 -5
qm 8 317.63578947 -5.435789473684 -10
9 326.52935673 1.270643274854
10 326.52935673 8.870643274854 -15
11 324.5554386 -5.155438596491
12 377.91684211 5.883157894737
13 351.23614035 -11.33614035088
14 351.23614035 -2.536140350877
15 342.3425731 3.757426900585
16 333.44900585 -6.249005847953
17 333.44900585 -0.549005847953
18 342.3425731 3.457426900585
19 360.1297076 3.170292397661
20 342.3425731 9.557426900585
246.49+8.89X1+15.81X2 CI Rooms
6.8647323082
e price will decrease, on average, by 8.89 $ b0: When there are no rooms and in an east 10.922402195
for each increase in rooms. neighbourhood, prices would be 246.49 $ 6.86<=b1<=10
.92

e price will be, on average, 15.81 $ greater

west side neighbourhood than an east side. F critical 2.64 81.84% of the variation in prices is explained
F STAT>2.64 variation of rooms and neighbourhoo
Reject H0
At least one independent variable affects Y
MS F Significance F t STATs>1.74 Reject H0
3054.9695614 43.8020651088 1.961782039E-07 There's evidence that both rooms &
neighbourhood affect
69.744875129 prices at 95%

t Stat P-value Lower 95% Upper 95% Lower 95,0% Upper 95,0%
23.018668355 2.97395286E-14 223.89503913839 269.07946379 223.89503914 269.07946379
7.6274351786 6.94368286E-07 6.43352841922 11.353606084 6.4335284192 11.353606084
4.1833228956 0.00062367504 7.837984465416 23.788448283 7.8379844654 23.788448283

t critical 1.74 r(y2|1)^2

r(y1|2)^2
Residuals 64.78% of prices are explained by rooms
20 number and 40.17% are explained by
neighbourhood.
15

10 The regression model is the most

appropriate
5

0
0 5 10 15 20 25
-5

-10

-15

The assumptions aren't valid

 SUMMARY OUTPUT
CI neighb
9.23591 Regression Statistics
22.390523 Multiple R 0.8186472
9.24<=b2<
=22.39 R Square 0.6701833
Adjusted R 0.6518601
ation in prices is explained by the Standard E 11.561938
rooms and neighbourhood. Observatio 20

ANOVA
df SS MS F Significance F
Regression 1 4889.3905 4889.3905 36.57576666726 1.022068596E-05
Residual 18 2406.2115 133.67842
Total 19 7295.602

Coefficients
Standard Error t Stat P-value Lower 95%
Intercept 247.51037 14.82093 16.700057 2.111213646E-12 216.3727526602
X Variable 9.6458611 1.5949398 6.0477902 1.022068596E-05 6.29501699471

SUMMARY OUTPUT
0.6477527
0.4017389 Regression Statistics
e explained by rooms Multiple R 0.5303882
% are explained by R Square 0.2813117
Adjusted R 0.2413845
Standard E 17.067292
model is the most Observatio 20
ropriate
ANOVA
df SS MS F Significance F
Regression 1 2052.338 2052.338 7.0456273039084 0.016139449048
Residual 18 5243.264 291.29244
Total 19 7295.602

Coefficients
Standard Error t Stat P-value Lower 95%
Intercept 325.64 5.3971515 60.335531 3.141202959E-22 314.3010054337
X Variable 20.26 7.6327249 2.6543601 0.016139449048 4.224240100709
Upper 95%Lower 95,0%
Upper 95,0%
278.64799 216.37275 278.64799
12.996705 6.295017 12.996705

Upper 95%Lower 95,0%

Upper 95,0%
336.97899 314.30101 336.97899
36.29576 4.2242401 36.29576
Success in MBA Program Undergraduate GPA GMAT Score Predicted probability Odds Ratio
0 2.93 617 0.00079058476731306 0.00079121
0 3.05 557 0.00114149941159357 0.0011428
0 3.11 599 0.00088039280881099 0.00088117
0 3.24 616 0.00079100271659036 0.00079163
0 3.36 594 0.00090381218260143 0.00090463
0 3.41 567 0.00106633834220818 0.00106748
0 3.45 542 0.00124288361164249 0.00124443
0 3.6 551 0.00117268853280594 0.00117407
0 3.64 573 0.00102337586020594 0.00102442
0 3.57 536 0.00128684630523033 0.0012885
1 2.75 688 0.00051217103294853 0.00051243
1 2.81 647 0.00065862708281407 0.00065906
1 3.03 652 0.00063610221398258 0.00063651
1 3.1 608 0.00083307082895553 0.00083377
1 3.06 680 0.00053502568720593 0.00053531
1 3.17 639 0.0006873879192421 0.00068786
1 3.24 632 0.00071676423369058 0.00071728
1 3.41 639 0.00068439752361214 0.00068487
1 3.37 619 0.00077468897493997 0.00077529
1 3.46 665 0.00058257631445312 0.00058292
1 3.57 694 0.00048628375428684 0.00048652
1 3.62 641 0.00067344329144415 0.0006739
1 3.66 594 0.00089890101138871 0.00089971
1 3.69 678 0.00053549437117328 0.00053578
1 3.7 624 0.00074670427451749 0.00074726
1 3.78 654 0.00061981003893106 0.00062019
1 3.84 718 0.00041738536165881 0.00041756
1 3.77 692 0.00049052730284402 0.00049077
1 3.79 632 0.00070963863564608 0.00071014
1 3.97 784 0.00027725658051286 0.00027733
 SUMMARY OUTPUT

a Regression Statistics ln(oddsratio)=-3.29+0.02*X1+0.01*X2

b Multiple R 0.7200343561
c R Square 0.5184494739 b0: the log-odds of success in the MBA
program when both the Undergraduate
Adjusted R Square 0.4827790646 GPA&GMAT score are zero is -3.29.
Standard Error 0.3448207715
Observations 30

ANOVA
df SS MS F
Regression 2 3.456329826247 1.7281649131 14.534441392
Residual 27 3.210336840419 0.1189013645
Total 29 6.666666666667

Coefficients Standard Error t Stat P-value

Intercept -3.2853130146 0.905894529364 -3.62659549 0.001177908
Undergraduated GPA 0.0181785769 0.201349083393 0.0902838822 0.9287280722
GMAT Score 0.0061642958 0.00117004937 5.2684066133 1.481731E-05
-3.29+0.02*X1+0.01*X2 b1: The log-odds of success for a one-unit
increase in Undergraduate GPA increases by
0.02, holding the GMAT score constant.
f success in the MBA
th the Undergraduate
are zero is -3.29.
b2: The log-odds of success for a one-unit
increase in GMAT Score increases by 0.01,
holding the Undergraduate GPA constant.

Significance F 0.0008727652
5.195952E-05 the probability of success of the program for a
student with an undergraduate GPA of 3.25 and a
GMAT score of 600 is 0.001

Lower 95% Upper 95% Lower 95,0% Upper 95,0%

-5.1440550547 -1.4265709746 -5.1440550547 -1.4265709746
-0.3949556168 0.4313127707 -0.3949556168 0.4313127707
0.00376355284 0.0085650388 0.0037635528 0.0085650388