QUESTION 1 (50 marks)

The file SAVINGS_2019 contains data from a hypothetical economy (quarterly; 2005:1 through 2018:4) on the following variables:
t:
q:
s:
y:
r1:
r2:
midage:

Time trend
Quarter (1 if JFM; 2 if AMJ; 3 if JAS; 4 if OND)
Real savings (£billion at 2018 prices)
Real personal disposable income (£billion at 2018 prices)
Short term interest rate (%)

Long term interest rate (%)
Percentage of the population in the 30-60 age group

(a) What were short term and long term interest rates in the second quarter of 2007?
The short term interest rate in the second quarter of 2007 was 2.00%.
The long term interest rate in the second quarter of 2007 was 2.76%.

(b) Using the command regress s y r1 r2 midage, estimate a regression model with s as the dependent variable, and y, r1, r2 and midage as explanatory variables. Report the coefficients and standard errors in the following table (USE 4 DECIMAL PLACES). Also report the R2 and the sample size.

(c) What is your estimate of the marginal propensity to save (MPS)? [Note: MPS is just the coefficient of y in the regression.] Find a 95% confidence interval for MPS. Interpret this interval.
The estimate of the marginal propensity to save (MPS) is 0.2431.
A 95% confidence interval for MPS ranged from 0.0876 to 0.3986.
We are 95% confident that the true MPS is somewhere between 0.0876 and 0.3986.

(d) Are either of the two interest rates important in explaining savings behaviour, according to tests of individual significance? (Hint: look at the t-statistics and p-values.)
Short term interest rate: β = 2.5781; t = 1.222, p = 0.2274
Long term interest rate: β = 0.0598; t = 0.032, p = 0.9748
The p-value for short term interest rate was 0.2274 > 0.05, so we fail to reject the null hypothesis that there is not an effect of short term interest rate on saving behaviour.
The p-value for long term interest rate was 0.9748 > 0.05, so we fail to reject the

4. Use the Park test to test the null hypothesis of homoskedasticit...