Kevin works as an analyst in an investment bank, and he tries to build a time series model for monthly U.S. inflation (denoted ) from Feb. 1971 to Dec. 2000. He first tried the Autoregressive AR(1) modelusing the previous month's inflation as the independent variable:

The table below shows the results of estimation.

AR(1) model: U.S. Monthly Inflation Feb. 1971 to Dec. 2000

Regression Statistics

R Square0.3808

Standard Error3.4239

Observations359

Durbin-Watson2.3059

Coefficients

Standard Error

t Stat

Intercept ( )

1.9658

0.2803

7.0119

Lag1 ( )

0.6275

0.0410

15.3049

Autocorrelation of the Residual

Lag

Autocorrelation

Standard Error

t Stat

1

-0.1538

0.0528

-2.9142

2

0.1097

0.0528

2.0782

3

0.1657

0.0528

1.2442

4

0.10920

0.0528

1.7434

Based on regression results in the table, discuss whether the estimates of and are valid, and give reasons. If the model is misspecified, describe what is the next step you should take to determine an appropriate Autoregressive time series model.