meta user
network profile
| bio | website | |
|---|---|---|
| location | San Jose, CA | |
| age | 33 | |
| visits | member for | 1 year, 7 months |
| seen | 10 mins ago | |
| stats | profile views | 334 |
Mad data scientist.
|
1d |
comment |
How do you explain 'co-integration' to determine spurious regression to a fairly new time series student? To quote xkcd 552, "Correlation doesn't imply causation, but it does waggle its eyebrows suggestively and gesture furtively while mouthing 'look over there'." |
|
1d |
revised |
How do you explain 'co-integration' to determine spurious regression to a fairly new time series student? added 58 characters in body |
|
1d |
revised |
What is the meaning of “All models are wrong, but some are useful” added 104 characters in body |
|
1d |
answered | What is the meaning of “All models are wrong, but some are useful” |
|
2d |
answered | How do you explain 'co-integration' to determine spurious regression to a fairly new time series student? |
|
May 14 |
answered | AIC, BIC parsimony |
|
May 10 |
comment |
Where to find datasets related to US elections at the individual level? This is not the right forum for this question. Try area51.stackexchange.com/proposals/37195/data |
|
May 10 |
comment |
Statistical package that works with sql database I would take a serious look at Stata. It is certainly beginner-friendly in that it has a GUI for many of the built-in commands, but the GUI also spits out the actual code so that you can progress to Jedi levels. It also has excellent pdf documentation. |
|
May 10 |
revised |
Statistical package that works with sql database added 291 characters in body |
|
May 10 |
answered | Statistical package that works with sql database |
|
May 10 |
comment |
Today's popularity of main data mining and machine learning tasks Rexer Analytics has done a survey about methods and software every year since 2007. They ask about particular algorithm folks use, so you may be able to get a time series from that. Here's the link to the most recent one: rexeranalytics.com/Data-Miner-Survey-2013-Intro.html |
|
May 9 |
comment |
Obtaining base level with margeff & gologit2 in STATA I somehow missed that! Dr. Williams' answer (stata.com/statalist/archive/2013-04/msg00950.html) avoids all my nonsense with preserve/replace/predict/restore, but the basic idea is the same. |
|
May 9 |
answered | Obtaining base level with margeff & gologit2 in STATA |
|
May 9 |
accepted | multiplicative treatment effects with standard errors |
|
May 9 |
accepted | How to capture competitive spatial interactions between multiple stores and customers |
|
May 9 |
comment |
How can a vector of variables represent a hyperplane? @Scott Yes. I think the logic applies to a higher number as well, but it would be hard to draw. |
|
May 9 |
comment |
Comparison of transformed predictors In response to your clarification, let me correct this common misconception. OLS does not require your right-hand-side, explanatory variables to be normally distributed. Transformations of RHS variables may be helpful for theoretical reasons or analytical convenience, but least squares will work just fine without them. |
|
May 8 |
comment |
Comparison of transformed predictors You might want to add why you want to transform these explanatory variables. They do not need to be normally distributed for OLS. |
|
May 8 |
answered | Comparison of transformed predictors |
|
Apr 30 |
revised |
Why is there a sharp elbow in my ROC curves? added 20 characters in body |
