Scribd
Upload
10 views3 pages

Econometrics_Problem_Set_2

The document contains an econometrics problem set with three questions focusing on regression analysis and hypothesis testing. It includes detailed calculations and interpretations of coefficients, standard errors, fitted values, residuals, and R-squared values. The results indicate significant relationships between variables such as income and expenditure, as well as sales and CEO salary.

Uploaded by

Kiệt Đỗ
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
10 views3 pages

Econometrics_Problem_Set_2

The document contains an econometrics problem set with three questions focusing on regression analysis and hypothesis testing. It includes detailed calculations and interpretations of coefficients, standard errors, fitted values, residuals, and R-squared values. The results indicate significant relationships between variables such as income and expenditure, as well as sales and CEO salary.

Uploaded by

Kiệt Đỗ
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Econometrics Problem Set 2

Do Binh Kiet - 220251

October 10th , 2024

1 Question 1
yi = β̂0 + β̂1 xi + ϵi
c1 yi = β̂‘0 + β̂1′ c2 xi + ϵ′i
Replacing yi with β̂0 + β̂1 xi + ϵi , we have:

c1 (β̂0 + β̂1 xi + ϵi ) = β̂‘0 + β̂1′ c2 xi + ϵ′i


c1 β̂0 + c1 β̂1 xi + c1 ϵi = β̂‘0 + β̂1′ c2 xi + ϵ′i
Equating the coefficients of the independent variable −→ c1 β̂1 = β̂‘1 c2
c1
β̂1′ = β̂1
c2
Equating the constants −→ c1 β̂0 = β̂‘0
β̂0′ = c1 β̂0

Alternatively, we can derive this result using the formula for β̂1 and β̂0
Pn
′ (c2 xi − c2 x̄)(c1 yi − c1 ȳ)
βˆ1 = i=1Pn
(c2 xi − c2 x̄)2
Pni=1
c1 c2 i=1 (xi − x̄)(yi − ȳ)
= 2 Pn 2
c2 i=1 (xi − x̄)
c1 ˆ
= β1
c2

With this result, we can compute β̂0′ as:


′
β̂0′ = c1 ȳ − c2 βˆ1 x̄
c1
= c1 ȳ − c2 βˆ1 x̄
c2
= c1(ȳ − βˆ1 x̄)
= c1β̂0

1
2 Question 2
(a) We have the following OLS regression results:

Figure 1: OLS Regression results table

From this result, we can obtain the following equation for the relationship
between Expenditure and Income: Y = 0.6730X + 1.8486. The standard
error for the coefficient of X is 0.042 and the standard error for the constant
is 0.584.
(b) The coefficient 0.6730 indicates that for 1 unit of increase in income, ex-
penditure increases by 0.6730 unit. This suggests that expenditure does
not increase 1-for-1 with income but instead people have a tendency to
divert some of the extra unit of income to savings. The constant 1.8486
indicates that if income is 0 then expenditure will be 1.8486, which reflects
the reality that even those who do not make any income still have to find
ways to spend money to survive. With the p-value for income being 0
and the p-value for the constant being 0.013, the regression results are
significant at a 5% significant level.
(c) To calculate the fitted values, we plug the income variable (X) in the
equation and solve for expenditure (Y). The residuals can be obtained by
subtracting the fitted values from the observed values for every observa-
tion. Thus, we have the following table:

2
Figure 2: Table with fitted values and residuals

(d) R-squared = 0.969 means that 96.9% of the variability in Expenditure can
be explained by changes in Income. This suggests a strong relationship
between the two variables.

3 Question 3
(a) For a firm with 0 sales and 0 log(price), log(salary) will be 4.32. This
means that when the firm sells nothing and its stock price stays the same
then the salary for its CEO is predicted to be e4.32 ≈ 75.1886(units). An
1-million increase in sales is associated with a 0.08 increase in log(salary).
In other words, CEO’s salary will be multiplied by e0.08 ≈ 1.0833 for every
1-million increase in sales. 1 unit of increase in log(price) is associated with
0.174 unit of increase in log(salary). In other words, as stock price rises
by 1%, CEO’s salary will increase by 0.174%.
(b) Let β1 be the coefficient for sales:
H0 : β1 = 0
H1 : β1 > 0
Let the standard error for β1 be se1
β1 0.08
The test statistic is t = se1
= 0.025 = 3.2
With a sample size of 209 and 2 independent variables, the degree of
freedom for this test is 209 − 2 − 1 = 206
Thus, given a 5% significant level, the critical value will be approximately
1.6523, which is smaller than 3.2.
Therefore, we reject the null hypothesis. There is sufficient proof that
sales positively predict CEO’s salary at a 5% significant level.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy