Academic ECONOMETRICS WRITER USING EVIEWS
Budget: £20 – £250 GBP
Exercise 1Consider the following model:votet = 0 + 1g row tht + 2goodnewst + 3wart + utwherevote = percentage share of the popular vote won by the incumbent party.1growth = annual growth rate in real per capita GDP in the Örst 15 quartersof an administration.goodnews = number of quarters in the Örst 15 quarters of the administrationin which the growth rate of real per capita GDP is greater than3.2% at an annual rate.war = dummy variable equal to 1 for the elections of 1920, 1944 and 1948,and zero otherwise.1. Estimate the model, write out the estimated equation and provide the interpretationof all coe¢ cient estimates.2. Perform an appropriate test for autocorrelation of order one in ut.3. Test whether the growth rate has a di§erent e§ect on vote for election years 1920, 1944and 1948 and the rest of the years in the study?4. Test for misspeciÖcation using the Ramsey RESET set.Notes: (1) The dataset for this exercise, called "data_elections.wf1 ", contains data for theUnited States presidential elections years from 1880 to 2008 (31 observations). (2)Describe the implementation of all tests in detail step by step. (3) For each questionprovide the relevant EViews outputs or/and plots.
Exercise 2
Consider the analysis of quarterly data, from 1980 to 2018, of the variables GDPt (income), CAPt (stock of capital) and LABt (stock of labour).
1. Apply the Engle and Granger (EG) procedure to test whether the model below constitutes a cointegrating relationship.
GDPt= 0+ 1CAPt+ 2LABt+"t 1Incumbent means the party in power at the time of the election.
2
Use the Augmented Dickey Fuller (ADF) test ó with intercept and no time trendó for stationarity in all cases. Detail the auxiliary regression, null/alternative hypothesis, as well as all steps and conclusion for each of the ADF tests employed.
2. Explain the consequences of the result of the EG procedure employed in part 1 for the reliability of the regression above.
3. Respecting the error correction model (ECM) for the model above:
3.1 Specify the equation of ECM, estimate the model and provide the estimated ECM equation.
3.2 Interpret the adjustment parameter estimate of the ECM?
4. Explain the di§erence between the Dickey-Fuller (DF) and Augmented Dickey-Fuller (ADF) tests for stationarity? You should use equations to illustrate your answer.
5. Why is the ADF test generally preferred to the DF test? Explain your answer.
Notes (1) The dataset for this exercise is "gdp.wf1". (2) Describe the implementation of all tests in detail step by step. (3) For each test provide the relevant EViews outputs or/and plots.
Exercise 3
Consider the following simultaneous equations model (SEM)
Pt = 0+ 1Wt+ 2It+ 3imt+ut Wt = 0+ 1Pt+ 2Ut+ 3Xt+ut
where P is the rate of growth of prices, W is the rate of growth of wages, I is the rate of growth of investment, U is the rate of unemployment, im is the rate of growth of imports and X is the rate of growth of productivity.
Answer all questions below:
1. Classify the variables in the model above according to whether they are endogenous or exogenous.
2. Derive the model reduced-form equations (RFE) and determine whether the equations are identiÖed.
3
3. Using the RFE, explain how the classical linear regression model assumption that ìthe error term is uncorrelated with the explanatory variablesîis not met, and discuss the consequences of this violation on the ordinary least squares (OLS) estimator.
4. Can indirect least squares (ILS) be used to estimate the parameters of this SEM? Why?
5. Explain the two-stages least squares (TSLS) estimation method for this SEM.
Exercise 4
1. Write out the following processes: ADL(1,1); ADL(0,3); DL(2) for the static equation below
Yt= + Ct+ It+Gt+"t:
2. Explain step by step how an AR(1) process can be derived from each of the following
three models: Koyck, Adaptive Expectations (AE) and Partial Adjustment (PAM).
3. - over limit - will send this sentence if you accept project
Exercise 2
Consider the analysis of quarterly data, from 1980 to 2018, of the variables GDPt (income), CAPt (stock of capital) and LABt (stock of labour).
1. Apply the Engle and Granger (EG) procedure to test whether the model below constitutes a cointegrating relationship.
GDPt= 0+ 1CAPt+ 2LABt+"t 1Incumbent means the party in power at the time of the election.
2
Use the Augmented Dickey Fuller (ADF) test ó with intercept and no time trendó for stationarity in all cases. Detail the auxiliary regression, null/alternative hypothesis, as well as all steps and conclusion for each of the ADF tests employed.
2. Explain the consequences of the result of the EG procedure employed in part 1 for the reliability of the regression above.
3. Respecting the error correction model (ECM) for the model above:
3.1 Specify the equation of ECM, estimate the model and provide the estimated ECM equation.
3.2 Interpret the adjustment parameter estimate of the ECM?
4. Explain the di§erence between the Dickey-Fuller (DF) and Augmented Dickey-Fuller (ADF) tests for stationarity? You should use equations to illustrate your answer.
5. Why is the ADF test generally preferred to the DF test? Explain your answer.
Notes (1) The dataset for this exercise is "gdp.wf1". (2) Describe the implementation of all tests in detail step by step. (3) For each test provide the relevant EViews outputs or/and plots.
Exercise 3
Consider the following simultaneous equations model (SEM)
Pt = 0+ 1Wt+ 2It+ 3imt+ut Wt = 0+ 1Pt+ 2Ut+ 3Xt+ut
where P is the rate of growth of prices, W is the rate of growth of wages, I is the rate of growth of investment, U is the rate of unemployment, im is the rate of growth of imports and X is the rate of growth of productivity.
Answer all questions below:
1. Classify the variables in the model above according to whether they are endogenous or exogenous.
2. Derive the model reduced-form equations (RFE) and determine whether the equations are identiÖed.
3
3. Using the RFE, explain how the classical linear regression model assumption that ìthe error term is uncorrelated with the explanatory variablesîis not met, and discuss the consequences of this violation on the ordinary least squares (OLS) estimator.
4. Can indirect least squares (ILS) be used to estimate the parameters of this SEM? Why?
5. Explain the two-stages least squares (TSLS) estimation method for this SEM.
Exercise 4
1. Write out the following processes: ADL(1,1); ADL(0,3); DL(2) for the static equation below
Yt= + Ct+ It+Gt+"t:
2. Explain step by step how an AR(1) process can be derived from each of the following
three models: Koyck, Adaptive Expectations (AE) and Partial Adjustment (PAM).
3. - over limit - will send this sentence if you accept project