Estimation of Parameters in Time‐Series Regression Models

摘要: We consider the estimation of coefficients in a general linear regression model which some explanatory variables are lagged values dependent variable. For discussing optimum properties concept best unbiased estimating equations is developed. It shown that when errors normally distributed method least squares leads to estimates. The least-squares estimates be same asymptotically as those ordinary models containing no variables, whether or not distributed. Finally, proposed for different has but have an autoregressive structure. efficient large samples.