You are on page 1of 3

Study Questions 2: Appication questions

1) The regression equation for demand for money for 8 different economies was
estimated by OLS as follows;

where Y is the demand for money ($billions) and X is interest rate (%). Given the
following data as

a) Compute the estimate of the error variance, .


b) Compute the estimate of the standard error of the slope coefficient,
c) Compute the t-value and test whether the slope coefficient is significantly
different from zero at the significance level of 0.05? State the null and the
alternative hypothesis.
d) It is claimed that 1% increase in interest rate will lead to a fall of 10 billion
dollars in the demand for money. Test this claim using significance level of
0.05. State the null and the alternative hypothesis.
e) What type of data is used for the above econometric model?

Answers:

a) = RSS/n-2 (RSS = )
= 130/8-2
= 21.67

b) = =

=
= 1.033

c)

Note that the t-statistics are computed under the Ho: This is equivalent to
dividing the estimated coefficient by the standard error of the estimated
coefficient as above.

Therefore, under the against since the calculated


t value = -5.26, and the critical t- value = 2.447 with df = n-2 = 6 at 5% level of
significance, we reject the null hypothesis and conclude that the slope coefficient
is significantly different from zero. This means that the X variable is a relevant
(significant) variable that belongs into the regression model.

d) against

since critical t = 2.447 we reject null

hypothesis and conclude that 1% increase in interest rate will not lead to a fall of
10 billion dollars in the demand for money.

f) Cross-section data.

2) The general manager of an engineering firm wants to know if technical worker’s


experience influences the quality of their work. A random sample of 24 workers is
selected and their years of experience and their quality rating (as assessed by their
supervisors) recorded. Work experience (EXP) is measured in years and quality rating
(R) take values between 1 trough 7 (with 7 = excellent and 1=poor) The least squares
estimates of the model and the standard errors of the estimates are

s.e (0.709) (0.044)

a) Interpret the intercept and the coefficient of EXP.


b) Construct a 90% confidence interval for the slope of the relationship between
quality rating and experience. Using the confidence interval you have constructed
test whether Ho : = 0 against the H1 : ≠ 0.
c) Test whether experience has a positive impact on quality rating using significance
approach at 10% level of significance.

Answers:

a) For the slope coefficient: An additional year of work experience will lead
to an average increase in quality rating by 0.076.Or, alternatively, you can also
say, on average, a technical artist’s quality rating goes up by 0.076 for every
additional year of experience.
For the intercept: When work experience is zero years, the average quality
rating will be equal to 3.2

b) so 90% confidence interval is


0.0005 and 0.152. In testing the above hypothesis, the value of zero falls outside
the interval which means that null hypothesis is rejected. So, we conclude that
work experience is a significant variable in the above regression model.
c) When using the significance approach, we need to calculate the t-statistic;

Ho: 0 against the H1: > 0.

The right-tail critical t-value for one-tail test is 1.321 for 90% confidence level.
(df=22). Since 1.727 > 1.321, we reject null and conclude that slope coefficient ( ) is
positive. Experience has positive effect on quality rating.

You might also like