Reputation
16,711
Next tag badge:
94/100 score
24/20 answers
Badges
5 27 63
Newest
 Nice Answer
Impact
~448k people reached

14h
comment Do you have to adhere to the likelihood principle to be a Bayesian?
xianblog.wordpress.com/2014/11/13/…
16h
comment How Do I choose parameters of prior on regression coefficients in a Bayesian linear model?
See elicitation.
2d
comment How to studentize residuals
Please remember to attribute material from elsewhere, even in questions.
2d
revised How to studentize residuals
added attribution
2d
comment Add a legend on package darts
Welcome to Cross Validated! This isn't on topic here, but with a minimal, complete, verifiable example might be on Stack Overflow.
2d
comment One to one exact matching in R
Welcome. If you run into problems check on Stack Overflow & ask there if there's not already an answer.
2d
comment Improve ctr for recommender system
Welcome to Cross Validated! This is rather a broad question for our site & I'd recommend focussing it on some specific aspect of recommender systems, perhaps after reading up on them. By the way, please expand initialisms such as CTR on first use.
2d
comment One to one exact matching in R
See ?merge - you shouldn't need much code.
2d
revised Boruta feature selection with R
appended answer 194566 as supplemental
2d
comment omit variable to find feature importance in classification
I imagine the idea behind permuting is to remove a variable's signal, leaving only noise. If you omit a variable you'll remove both signal & noise, so if it's not very predictive you might even find classification accuracy increases - when you're using bagging or the like.
2d
comment chi squared with proportions
Anyway, given that in Field 1 you've counted 650 molehills, what would the expected counts be if the chances of a molehill's occurring in each bin were equal?
2d
comment chi squared with proportions
Your null hypothesis shouldn't be "there is no difference in frequency of mole hills observed in each bin in Field 1"; you already know there's a difference in the observed frequencies. It should be that there's an equal chance of a molehill's occurring in any bin. (Which seems a little odd given that the last bin's any distance over 12m. How far from the fence did you look for molehills?) And you're probably assuming that each molehill's made independently of the others. Does that seem reasonable? - did you notice any clustering of molehills?
Feb
7
comment Do you have to adhere to the likelihood principle to be a Bayesian?
Compare those for the binomial & negative binomial.
Feb
7
comment Do you have to adhere to the likelihood principle to be a Bayesian?
Not necessarily. And I'm not sure what difference it makes.
Feb
7
comment Do you have to adhere to the likelihood principle to be a Bayesian?
No - see the Jeffreys prior. Bayesian methods may violate the (strong) likelihood principle.
Feb
5
revised Complete separation and stepwise regression - possible in R?
added 441 characters in body
Feb
5
answered Complete separation and stepwise regression - possible in R?
Feb
5
comment compare regression coefficients between subgroups
... or edit this question if there's a difference from the others that needs bringing out.
Feb
5
comment compare regression coefficients between subgroups
Thanks for adding more information. That seems exactly like the situation described in the post I linked to. Others worth reading are What test can I use to compare slopes from two or more regression models?, What is the correct way to test for significant differences between coefficients?, & How to test whether a regression coefficient is moderated by a grouping variable?. Have a look at these & post a follow-up question if you have one - ...
Feb
5
revised compare regression coefficients between subgroups
appended answer 194186 as supplemental