Timeline for How can I improve my analysis of the effects of reputation on voting?
Current License: CC BY-SA 3.0
9 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 17, 2011 at 0:50 | comment | added | cardinal | @StasK: That's an extremely weak definition of "stationarity" that you are using! For example, let $X_i$ be i.i.d standard normal and take $Z_n = n^{-1/2} \sum_{i=1}^n X_i$. The marginals of $Z_n$ are constant with respect to $n$ (indeed, standard normal!) but $\{Z_n\}$ is not stationary by any standard definition. | |
| Aug 11, 2011 at 15:36 | comment | added | StasK | A time series is stationary if the marginal distributions at all time points are the same. So even you might have the same mean, a changing variance will make the series non-stationary. An example are (G)ARCH models for which a Nobel prize was given in the early 2000s. But in these data, I would expect some shifts in the mean, as well. If the audience of the website grows, then for a given quality of an answer, you would likely to see more votes on it, which will likely raise both the mean and the variance of the scores. | |
| Aug 11, 2011 at 14:51 | comment | added | Andy W | I'm a bit confused by the last comment. How do exogenous factors that affect the score of an answer make the series heteroskedastic (I assume you mean that the variance of score gets larger/smaller with post number?), and of what relevance is this to the question at hand? | |
| Aug 11, 2011 at 13:24 | history | edited | StasK | CC BY-SA 3.0 |
some comments on issues with non-stationarity
|
| Aug 11, 2011 at 13:22 | comment | added | StasK | At the very least, it is probably heteroskedastic: some posts are interesting, get a lot of hits and a lot of upvotes, while others are small clarifications or RTFM-"Read this link" type of questions/answers. That of itself would technically make it non-stationary. Of course stationarity is a testable assumption, but with crazy data like these, you'd probably want to be on a very safe side of being overly conservative in the analysis methods (or, as I mentioned, to be aware that the results may be weird). | |
| Aug 11, 2011 at 12:52 | comment | added | Andy W | Also I highly doubt score is non-stationary, what makes you think it is? | |
| Aug 11, 2011 at 12:49 | comment | added | Andy W | thank you for the response, and a few comments/questions. I agree I should have at least explored a more explicit time series approach in this data (I did not even check to see if there was any evidence of autocorrelation in the residuals). There are a few more complications though in time series modelling of this data (what is t?, and score itself is dynamic and not fixed per post number), also there would be no need for a regression predicting Z_t, I know perfectly what Z_t is a function of! | |
| Aug 11, 2011 at 7:55 | history | edited | StasK | CC BY-SA 3.0 |
addressed another part of the original question
|
| Aug 11, 2011 at 7:30 | history | answered | StasK | CC BY-SA 3.0 |