In this rst part of the assignment you will t and interpret a frequentist VAR model for the data. Perform the following analysis and be sure to fully interpret and report your results (i.e., don’t just tell us the lag length for the VAR is p. Explain how you came to this conclusion and other results).
1. Test the series to determine the lag length for a VAR model.
2. Using the lag length you select in the previous step, conduct a Granger causalityanalysis. Report the results in a table.
3. Fit a reduced form VAR model for these data.
4. Generate impulse response functions (IRFs) for the reduced form model you t in thelast step. Given that the data are monthly and this is American political data, youwill need to make a choice about what time horizon to use.
5. Now redo the previous step and generate a set of Monte Carlo error bands for theimpulse responses using 10000 samples.
6. Now comment on what dynamic factors aect presidential approval and macroparti-sanship. That is, interpret the impulse responses you generated.
The link to the data is: https://www.dropbox.com/s/egcf0kkr4b561oq/polity.d…