In this assignment, you will have to perform dynamic analysis for macroeconomic data available from the Australian Bureau of Statistics (ABS). To determine the data set that you will have to analyse in this assignment, use the following data:
Frequency: Quarterly. Sample period: Use all available data. Source: Table 26 from http://www.abs.gov.au/AUSSTATS/abs@.nsf/DetailsPage/5206.0Jun%202016?OpenDocument
DATA SET D
: Natural log of public gross fixed capital formation in New South Wales, chain volume measure, seasonally adjusted.
: Natural log of real final (aggregate) demand for New South Wales, chain volume measure, seasonally adjusted.
Using your assigned data set, proceed as follows:
1. Investigate and describe how ABS has obtained the chain volume measures of the above variables.
2. Investigate and describe how ABS has perform the seasonal adjustment of the above time series.
3. Perform an Augmented Dickey Fuller (ADF) test to determine whether the series (in natural logarithmic form) are stationary or non-stationary. Explain this test very carefully.
4. Perform an ADF test to determine whether or not the series are cointegrated. Explain this test very carefully.
5. Based on the above tests, decide which of the time series models covered in class (the Engle-Granger error-correction model or the ARDL model, covered in Seminars 7-9) would be the most appropriate time series model to explain the relationship between the two variables.
6. Write the dynamic time series model selected in section 5. Interpret very carefully the coefficient estimators and the structure of the model.
7. Conduct appropriate causality tests. Explain carefully the hypotheses in each test and your conclusion.
8. Offer an overall interpretation of your results, emphasising findings that may be useful from a policy perspective.