Examining model selection criteria for single variable time series

In this study, several model selection criteria used in selecting appropriate model degree in single variable time series are examined. By using Akaike Information Criteria (AIC), Final Prediction Error (FPE), Hannan-Quinn Information Criteria (HQ), Adjusted Akaike Information Criteria (AICC) and Schwarz Information Criteria (SIC), simulations are performed to compare the performances of these criteria in selecting appropriate lag length for different model structures and sample size. In order to compare these criteria, Monte Carlo simulation method is employed to compute the lag lengths selected by each criteria. Findings of this simulation are submitted.