I am reading a journal where it is written "Time series are based on annual-mean translation speeds from 1949-2016. Trends are estimated by linear regression. The P values of the regression are based on the full degrees of freedom in the 68-year time series. Because some of the time series exhibit autoregressive (AR(1)) persistence, as determined from a Durbin–Watson test, confidence intervals are provided with degrees of freedom adjusted when needed. Statistical significance is based on the two-sided 95% confidence intervals (not the P values)". I am struggling with understanding the following question:

What is the relation between Autoregressive persistence and adjusting the confidence interval? Since I have a similar set of data set how can I adjust confidence interval when needed?



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.