serial correlation


Also found in: Dictionary, Medical, Financial, Wikipedia.
Related to serial correlation: multicollinearity

serial correlation

[′sir·ē·əl ‚kär·ə′lā·shən]
(statistics)
The correlation between values of events in a time series and those values ahead or behind by a fixed amount in time or space or between parts of two different time series.
References in periodicals archive ?
Therefore, the regression diagnostics tests have been used to check problems of heteroskedasticity and serial correlation in panel dataset used in this study for analysis.
Indeed, our simulations demonstrate large bias only when the matching variable has modest serial correlation and the difference in level between the two groups is large and stable.
(6) Additionally, the fact that both the independent and dependent variables exhibit serial correlation can result in false rejection of the null hypothesis that laws have no effect on a given outcome.
Different from Table A4, the estimation results in Table A5 pass the Hansen test for over-identification and Arellano-Bond test for serial correlation. Hence, we use the results in Table A5 as the main findings of our analysis.
Stability of the constructed model is tested on the basis of the departure of serial correlation and heterogeneity of the residuals under the null.
Result of Wooldridge test for serial correlation and modified Wald test for group-wise Heteroscedasticity show that there is no issue of autocorrelation and heteroscedasticity.
The diagnostic tests are used for finding serial correlation, functional form, normality and Heteroscedasticity among the variables of the model.
After conducting residual diagnostic tests, it was found that model 3 has problems of serial correlation and multicollinearity.
Again, the Durbin-Watson statistic does not indicate first-order serial correlation, and Q statistics from a correlogram of residuals do not indicate higher-order serial correlation.
We combine ordinary least square estimation as the first step, and generalized methods of moments as a second step, to guarantee that estimators are robust to serial correlation and heteroskedasticity.
Lagrange multiplier test (LM test) of residual serial correlation shows that there is no serial correlation among the variables as probability value is greater than 0.05.
Full browser ?