Unformatted text preview:

Economics 674 Fall 2006 Project 3 (Due: Friday, December 1) Let πt denote the inflation rate in period t. In the following problems, use the same inflation rate data you used in Projects 1 and 2 and the sample period 1960:I-2006:I. Report all pertinent results and accompany results with clear explanations and discussion. 1. Base Model Fit the inflation rate to an AR(p) model, where p is selected by applying the AIC with pmax = 4. 2. Test for Structural Change with an Unknown Breakpoint Test whether the parameters of the AR(p) model are constant over time by computing Quandt’s Likelihood Ratio (QLR) statistic and comparing it to the appropriate limiting distribution provided in Andrews (1993). (Set rmin = 0.1and rmax=0.9.) 3. Fit the Inflation Rate to a Threshold Model Fit πt to a two-regime self-exciting threshold model, assuming that the threshold is equal to the inflation rate’s sample mean. (Use the same p you selected for the base model. Choose the delay parameter, d, to minimize the SSR.) 4. Testing for ARCH Effects Apply Engle’s LM test to the residuals from the AR(p) model you fit in (1) to test the null hypothesis that the inflation rate is homoskedastic. (Choose the order of the AR regression for the squared residuals by applying the AIC with a maximum lag length of 4.) 5. Testing for a Unit Root Apply the ADF test to test the null hypothesis that there is a unit root in the inflation rate under each of the following assumptions: i. Under H0, the inflation rate is an I(1) process without drift. Under HA, the inflation rate is I(0) without any restriction on its mean. ii. Under H0, the inflation rate is an I(1) process with drift. Under HA, the inflation rate is trend stationary. Use the same p you selected in (1). 6. Cointegration Let it denote the nominal interest rate in period t. Using the same interest rate data you used in Projects 1 and 2 and use the sample period 1960:I-2006:I .Assume that πt and it are I(1) processes. i. Test the null hypothesis that that the real interest rate is I(1), with no drift, applying the ADF test. Select the lag length for the ADF regression by applying the AIC with pmax = 4. ii. Test the null hypothesis that πt and it are cointegrated by applying the Engle-Granger residual-based ADF test using a lag length of four in the second-stage


View Full Document

ISU ECON 674 - problem_set_3

Download problem_set_3
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view problem_set_3 and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view problem_set_3 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?