# ESTIMATION AND INTERPRETATION OF MULTIPLE REGRESSION MODEL

The Estimated Regression Equation is:

Production Function Data for the
Mexican economy
GDP
(Y)
11404
3
12041
0
12918
7
13470
5
13996
0
15051
1
15789
7
16528
6
17849
1
19945
7
21232
3
22697
7
24119
4
26088
1
27749
8
29653
0
30671
2

LABOR
(X2)

CAPITAL
(X3)

8310

182113

SUMMARY OUTPUT

8529

193749

Regression Statistics

8738

205192

Multiple R

8952

215130

ˆ +β
ˆ X +β
ˆ X
Yˆ = β
1
2 2
3 3

0.99768

9171

225021

R Square
Square

0.99536

9569

237026

Standard Error

5946.63957

9527

248897

Observations

20

9662

260661

10334

275466

ANOVA RESULT SUMMARY

10981

295378

df

11746

315715

Regression

2

11521

337642

Residual

17

11540

363599

Total

12066

391847

12297

422382

12955

455049

13338

484677

R
0.99482

SS
12903862451
0

MS
6451931225
5

F

Significance F

1825

0.000000000000000000015

35362522

19

601162876
12963978738
6

Coefficients

Standard Error

t Stat

P-value

Lower 95%

Upper
95%

Intercept

-38831.48

15422.34187

-2.5179

0.022

-71369.83

-6293.14

LABOR

7.46

2.59834

2.8728

0.011

1.98

12.95

32903 0 35405 7 37497 7 13738 520553 15924 561531 14154 609825 CAPITAL 0.0287 0.51 units holding labor input constant. which is the vertical intercept of the estimated regression line indicates that if capital and labor are zero.e.51X 3 ˆ . the value of GDP would be β 1 ˆ = -38831. An R2 of .00000000097 0.5% of the sample variability in GDP(Y) can be explained by its linear dependence on labor ( X 2 ) and capital ( X3) .831. Q-4: Are labor(X2) and capital (X3) significant independent variables or do they have any ability whatsoever to influence the dependent variable GDP (Y)? . GDP will change by 7. p-value = 1. β 2 ˆ = β 3 .48 also shows the average effect of all the other independent variables not included in the estimated regression model.995 shows that about 99.5*10-17 < 1%.46 implies that if labor is changed by 1 unit.46 units holding capital input constant.48. This suggests that the estimated goodness of fit is statistically significant. Q-3: Explain the ANOVA results or “Is the goodness of fit statistically significant”? ANOVA tests the following hypothesis: H 0 : R 2 = 0 → H 0 : β1 = β 2 = β 3 = 0 H1 : R 2 ≠ 0 → H1 : At least two betas are not equal to zero.60 Q-1: Write your estimated regression equation and explain it. The estimated regression equation is: Yˆ = −38831. we reject H 0 : R 2 = 0 .04256 12. GDP will change by .e. β 1 ˆ = 7. R2.46 X 2 + .48 + 7. Since the calculated F-statistic is statistically significant i.51 implies that if capital is changed by 1 unit. -38.42 0. Q-2: Explain the goodness of fit of your estimated regression line i.51 0.

7*10-9 respectively. As the regression summary suggests.42 < β 3 < . . which are 2. This is because p-values of β 2 and β 3 . with 95% confidence we can say that 1. are statistically ˆ and β Yes. As the regression results suggest. This significance tests are based on the hypotheses that: H 0 : β2 = β0 = 0 and H 0 : β3 = β0 = 0 .95 and that .87 and 12.011 and 9.98 < β 2 < 12.05. This implies that labor and capital have significant ability to influence the GDP of Mexico.03 respectively. None of the intervals include β0 = 0 implying a 95% chance that the actual beta coefficients β 2 and β 3 would not be zero. which are .60. The significance of the betas is also evident from the 95% confidence intervals for β 2 and β 3 . Q-5: Explain the 95% confidence interval for the betas. are both less than . the t-ratios associated with β 3 2 ˆ ˆ significant at 5% significance level.ˆ .