Timeline for OLS with Time Series Data - yay or nay?
Current License: CC BY-SA 3.0
10 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 19, 2019 at 13:50 | comment | added | Richard Hardy | @Confounded, I am not entirely sure, but I think that MLE is also biased. Under a normality assumption for the errors, MLE should coincide or nearly coincide with OLS estimator for AR and VAR models. There should be some literature on the topic, but I am not ready to look it up this time. | |
| Dec 19, 2019 at 12:51 | comment | added | Confounded | My understanding is that in such cases a Maximum Likelihood (or quasi-likelihood) estimator should be used instead, no? | |
| Dec 19, 2019 at 12:49 | comment | added | Richard Hardy | OK, if you only want a brief answer, then yes, the estimator is biased. (It is still consistent, i.e. asymptotically unbiased, and asymptotically normal.) A follow up question would be, is there a better estimator? Shrinkage estimators are usually good in terms of mean squared error and similar metrics, but they are typically even more biased in small samples. | |
| Dec 19, 2019 at 12:32 | comment | added | Confounded | Because my question (or rather, point) is directly related to the OP, in my opinion. I believe, the bias of OLS in AR is a well-know issue. An demonstration of the phenomenon is given, for example, here: Asatoshi Maeshiro (2000) "An Illustration of the Bias of OLS for $Y_t = λY_{t-1} + U_t$", The Journal of Economic Education, 31:1, 76-80. | |
| Dec 18, 2019 at 19:51 | comment | added | Richard Hardy | @Confounded, why don't you post it as a separate question? | |
| Dec 18, 2019 at 19:25 | comment | added | Confounded | But isn't the assumption of strict exogenity violated in, say, AR model and doesn't this then mean that OLS estimate is biased? Thank you | |
| Jun 26, 2017 at 18:55 | history | edited | Richard Hardy | CC BY-SA 3.0 |
edited body
|
| Jun 26, 2017 at 18:55 | comment | added | Richard Hardy | @Kuma, thanks! A lot of credit goes to Matthew Gunn and Christoph Hanck, of course. | |
| Jun 26, 2017 at 18:32 | vote | accept | Kuma | ||
| Jun 25, 2017 at 19:45 | history | answered | Richard Hardy | CC BY-SA 3.0 |