SerialCorrelation[Optional;Verybriefoverview].pdfVIP

Serial Correlation [Optional; Very brief overview] Richard Williams, University of Notre Dame, /~rwilliam/ Last revised January 22, 2015 [NOTE: These notes draw very, very heavily from Pindyck Rubinfeld’s Econometric Models and Economic Forecasts (1991 edition).] Introduction. I’ve never had much reason to worry about serial correlation in my work. But, if you are working with data that are collected repeatedly across time (something which I suspect is much more common in Economics than in Sociology) this may be a concern for you; or, you may read other work in which serial correlation is an issue. I’ll therefore briefly discuss the problem, and you can turn to Pindyck and Rubinfeld or other sources if you need to know more. What Serial Correlation is. When error terms from different (usually adjacent) time periods (or cross-section observations) are correlated, we say that the error term is serially correlated. Serial correlation occurs in time-series studies when the errors associated with a given time period carry over into future time periods. For example, if we are prediciting the growth of stock dividends, an overestimate in one year is likely to lead to overestimates in succeeding years. There are different types of serial correlation. Withfirst -order serial correlation, errors in one time period are correlated directly with errors in the ensuing time period. (Errors might also be lagged, e.g. if data are collected quarterly, the errors in Fall of one year might be correlated with the errors of Fall in the next year.) With positive serial correlation , errors in one time period are positively correlated with errors in the next time period. Consequences of Serial Correlation. Serial correlation will not affect the unbiasedness or consistency of OLS estimators, but it does affect their efficiency. With positive serial correlation, the OLS estimates of the sta