Next week I'm going to have a test reviewing time series. Our professor gave us about 100 true/false questions to review as a basis for creating flash cards. There are a few I have questions about which I was hoping someone could help review. Here are the questions and what I believe to be correct. Any help would be appreciated, especially for the one I'm completely lost on. Thank you

PACF for white noise is significant for short lags - T

ACF for random walk slowly decays to 0 - T

PACF and ACF are useful for identifying possible serial correlation in regression models if we apply to the residuals of OLS Model. - T

If AIC is lower than BIC we need to follow the BIC since it gives a better fit model - F

The box-pierce and Ljung-Box Q-statistics Q statistic tends to be larger when we have a white noise error. - F

Strong stationarity means the errors should have a normal distribution with a zero mean and constant variance - F

If a series is stationary reject the null hypothesis of the Dickey-Fuller test since the null of the DF test is stationary - F

The OLS is the best unbiased estimator if the residuals of the model show first order serial correlation since it shows the nonlinearity of the model. - F

Durbin-Watson statistics tend to be lower if the residuals from an OLS model shows white noise - Any answer I'd give here is pure guess not sure

Real GDP is a stationary series since it is very stable over time and less volatile - T


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.