Scribd
Upload
65 views

Tutorial 7

- The document presents regression results analyzing the relationship between car consumption (C), price (P), and income (Y) from 1995-2019. - It finds that the price elasticity of demand is statistically significant at the 10% level but income elasticity is not significant at the 5% level. - A Cobb-Douglas production function is estimated using output (Q), labor (L), and capital (K) data from 1989-2012. It finds that increases in L and K both have a positive effect on Q and are statistically significant at the 10% level. The regression model fits the data well with an R-squared value of 95.74%.

Uploaded by

HENG QUAN PAN
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
65 views

Tutorial 7

- The document presents regression results analyzing the relationship between car consumption (C), price (P), and income (Y) from 1995-2019. - It finds that the price elasticity of demand is statistically significant at the 10% level but income elasticity is not significant at the 5% level. - A Cobb-Douglas production function is estimated using output (Q), labor (L), and capital (K) data from 1989-2012. It finds that increases in L and K both have a positive effect on Q and are statistically significant at the 10% level. The regression model fits the data well with an R-squared value of 95.74%.

Uploaded by

HENG QUAN PAN
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 4

Question 1

The data from year 1995 to 2019, we obtained the following regression result (standard errors in
parentheses):

C= 2.58 – 0.48P + 0.65Y


(0.62) (0.18) (0.32)

where C = car consumption, per year


P = price for car
Y = real disposable income

(a) Is the price elasticity of demand for car statistically significant at 10% significant level?

H0: βp= 0
H1: βp≠ 0
t= -0.48-0/0.18= -2.6667
Critical value T0.10,22= +-1.717
T-test< critical value
Reject H0
Price elasticity demand for car is significant at 10 % significance level.

(b) Is the income elasticity of demand for car statistically significant at 5% significant level?

H0: βy= 0
H1: βy≠ 0
t= 0.65-0/0.32= 2.0313
Critical value T0.05,22= +-2.074
T-test< critical value
Failed to reject H0
Income elasticity of demand for car is not significant at 5% significance level.
Question 2

Table 3.1 gives data on annual real output (Q), input labor (L) and input capital (K) for
U.S. from 1989 to 2012.

Table 2 - Data on annual output (Q), labor (L) and capital (K)
year output labor capital
1989 100 100 100
1990 101 105 107
1991 112 110 114
1992 122 118 122
1993 124 123 131
1994 122 116 138
1995 143 125 149
1996 152 133 163
1997 151 138 176
1998 126 121 185
1999 155 140 198
2000 159 144 208
2001 153 145 216
2002 177 152 226
2003 184 154 236
2004 169 149 244
2005 189 154 266
2006 225 182 298
2007 227 196 335
2008 223 200 366
2009 218 193 387
2010 231 193 407
2011 179 147 417
2012 240 161 431

a. Given the data in Table 3.1, estimate the above Cobb-Douglas production function.
Dependent Variable: LNOUTPUT
Method: Leas t Squares
Date: 03/10/22 Time: 14:10
Sample: 1989 2012
Included observations: 24

Variable Coefficient Std. Error t-Statistic Prob.

C -0.177310 0.434293 -0.408272 0.6872


LNLABOR 0.807278 0.145076 5.564513 0.0000
LNCAPITAL 0.233053 0.063530 3.668415 0.0014

R-squared 0.957425 Mean dependent var 5.077336


Adjusted R-squared 0.953370 S.D. dependent var 0.269234
S.E. of regres sion 0.058138 Akaike info criterion -2.735511
Sum squared resid 0.070982 Schwarz criterion -2.588254
Log likelihood 35.82613 Hannan-Quinn criter. -2.696444
F-statistic 236.1219 Durbin-Watson stat 1.523452
Prob(F-statistic) 0.000000

lnoutput= -0.1773+ 0.8073 labor+ 0.2331 capital

b. Interpret the slopes.

When lnlabor increase by 1 percent, lnoutput increase by 0.8073 percent. Vice versa
When lncapital increase by 1 percent, lnoutput increase by 0.2331 percent. Vice versa

c. Which explanatory variables in estimated regression are statistically significant at 10%


significance level?

H0: βlnlabor= 0
H1: βlnlabor≠ 0

Prob.= 0.0000 (<0.1)


Significant at 10% significance level.
Reject H0.
Inlabor is significant at 10% significance level.

H0: βlncapital= 0
H1: βlncapital≠ 0

Prob.= 0.0014 (<0.1)


Significant at 10% significance level.
Reject H0.
lncapital is significant at 10% significance level.

d. Find the overall significance of the estimated regression at 1% significance level.

H0: βlnlabor= βlncapital= 0


H1: at least one βi≠ 0
(i= lnlabor, lncapital)

Prob.= 0.0000 (<0.01)


Siginificant at 1% significance level.
Reject H0.
The model is overall significant at 1% significance level.

e. Test at the 5% significance level the null hypothesis that the log of output does not depend
on the log of capital.

H0: βlncapital= 0 (lnoutput does not depends on lncapital)


H1: βlncapital≠ 0 (lnoutput depends on lncapital)

Prob.=0.0014 (<0.05)
Significant at 5% significance level.
Reject H0.
Lnoutput depends on lncapital.

f. How well is the estimated regression line fits the data?

R-squared= 0.9574
95.74% estimated regression line fits the data.

You might also like