SEC322 TUTORIAL SHEET FOR UNITS 2 AND 3
UNIT 2 TUTORIAL QUESTIONS
QUESTION ONE
Mafuso was tasked by his econometrics lecturer to estimate the relationship between money
supply (X) and inflation (Y) in Zambia for the period 1980 to 2017. She used the following
steps to model consumption function:
Step 1: Univariate analysis on the original variables.
Step 2: Univariate analysis of the differenced variables
Step 3a: He then regressing the original variables Y on X and obtained the following results;
Step 3b: Analysis of the residual from step 3a:
Step 4: Granger causality analysis;
1
Note: The 5% critical value is −3.37.
a. On the basis of the results under step 1 (use 5% critical value is −3.37):
i. Are the variable X and Y stationary or non-stationary? Explain.[4 marks]
ii. What is the significance of adding .[2 marks]
b. On the basis of the results under step 2 (use 5% critical value is −3.37), what is the
intergration order of X and Y? Explain.[4 marks]
c. On the basis of the results in steps 3a and 3b, would you conclude that the two
variables are cointegrated or spurious (use 5% critical value is −3.37)? Explain?[3
marks].
d. Considering your response to (c) what is the next step? [1 marks]
e. Explain your conclusion based on the results in step 4. (Use the 2-t rule of thumb
here) [4 marks].
QUESTION TWO
Chigubu was tasked by his econometrics lecturer to model consumption in Zambia. He used
log consumption (Y) and log GDP( X) for the period 1948 to 1984 and followed the
following steps to model consumption:
Step 1: He tested Y and X for unit root at 5 percent level of significance without a trend and
obtained the following results:
Yˆt 0.73 0.01Yt 1 R 2 0.04
(0.29)
Xˆ t 2.51 0.08 X t 1 R 2 0.08
(0.37)
Step 2: Taking the first difference of X and Y, and testing for unit root in differences he
obtained the following results;
Yˆt 12.06 1.41Yt 1 R 2 0.70
(8.29)
Xˆ t 25.63 1.10X t 1 R 2 0.52
(5.37)
Step 3: Regressing Y on X yields
Yˆt 10.45 0.95 X t R 2 0.95 d 0.5
(26.69)
Step 4: Testing the estimated residuals for unit root he obtained;
2
uˆ t 0.03 0.50u t 1 R 2 0.26
(1.38)
Step 5: Based on the results in step 4 he ran the error correction model and obtained;
Yˆt 0.30 0.16X t 0.12u t 1 R 2 0.95
(6.69) (1.89)
Note: The 5% critical value is −3.37. The values in the parentheses are calculated t-values.
a. On the basis of the results under step 1, is there a unit root in GDP and consumption?
Explain.
b. On the basis of the results above, would you conclude that the two variables are
cointegrated? Explain?
c. Based on the results in step 3 and step 4, would you say the consumption model is
spurious? Give reasons for your answer.
d. Considering the results in step 4, was step 5 necessary. Why or why not?
e. What is the order of integration of X and Y?
f. What is the name given to the coefficient attached to ut-1 in the equation under step 5?
g. Interpret the regression coefficients in the equations under steps 3 and 5.
QUESTION THREE
1. Consider the following stochastic process: . What is the mean of [3
marks]
2. Given that Y, X and W are time series variable. Specify the following regression models and explain the
meaning of the coefficients in each respective model;
a. AR(2) when Y is the dependent variable.
b. ARDL(1,1) when Y is the dependent variable and X is the independent variables.
c. DL(3), when Y is the dependent variable and X is the independent variables.
3
QUESTION FIVE
UNIT 3 TUTORIAL QUESTIONS
QUESTION ONE
4
QUESTION TWO
Mariana a student at Mulungushi University wished to investigate the factors that affect the
utilisation of modern and traditional health care services in Zambia. Using a sample of 1,6183
men and women, he ran a logistic regressed type of health care (categorised as modern = 1,
traditional = 0) on marital status (union; coded as married = 0, not married = 1),
employment status(employstatus; coded as 0 = unemployed, 1 = employed), v133 (education
in single years), v025 (coded as 1 = urban, 2 = rural). Using STATA 13.0, she obtained the
following results;
a. Interpret the odds ratios on the table above.
b. Comment on the significance of the estimates.