9,007 reputation
22572
bio website moderntoolmaking.blogspot.com
location San Francisco, CA
age 27
visits member for 3 years, 3 months
seen 8 hours ago

2d
comment Error metrics for cross-validating Poisson models
Thanks for adding R code. This is extremely useful.
Apr
14
reviewed Approve suggested edit on Understanding tail dependence coefficients
Apr
14
revised Data APIs/feeds available as packages in R
added 350 characters in body; edited tags
Apr
14
awarded  Notable Question
Apr
11
comment Implementation of online classification algorithms?
Not really. It's a C++ program you have to compile into a binary and run from the command line. I've written some wrapper functions around it in R to pass data to VW and read back the responses, but it's klunky. If you have a mac, you can install VW via homebrew.
Apr
9
comment Interpreting the “coefficient” output of the lm function in R
Lets say your average teacher rating variable is on a scale of 1-10. Re-scale it so it goes from 1-1000 by multiplying by 100 and you will see the effect described
Apr
9
comment Ensemble of models with different feature spaces
@user2932774 if you like my answer, please feel free to upvote or accept it! And yes, corrected your intercept term should work too.
Apr
9
revised should i take this variable which consist only 1 question in factor analysis? (all other variables are based on multiple questions)
added 2 characters in body
Apr
9
comment should i take this variable which consist only 1 question in factor analysis? (all other variables are based on multiple questions)
Hi Talha: I edited your question to have paragraphs, use complete sentences, and have proper capitalization. It still has some serious readability issues. I think you have a valid question in there somewhere, but you probably need a complete re-write to make it clear.
Apr
9
revised should i take this variable which consist only 1 question in factor analysis? (all other variables are based on multiple questions)
Serious formatting and grammar issues
Apr
9
answered Ensemble of models with different feature spaces
Apr
8
reviewed Approve suggested edit on R Interpret coefficient from Survreg(dist=“gaussian”)
Apr
3
comment Vowpal Wabbit: understanding prediction output
Good job answering your own question-- this explanation makes sense. Feel free to accept your own answer.
Apr
3
revised Vowpal Wabbit: understanding prediction output
edited tags
Apr
2
revised Cost functions for contextual bandits
Edits for clarity
Apr
2
comment Cost functions for contextual bandits
@alto: I think that should read "there are 2 possible responses we can record for a user". Does that make more sense?
Apr
2
asked Cost functions for contextual bandits
Apr
2
comment What time series model should be used?
You could also try the ets function and compare the results. Also try fit <- auto.arima(tsdata, ic = 'aicc', stepwise=FALSE)
Mar
26
reviewed Approve suggested edit on Normal curve probability mean
Mar
26
revised Compute Shannon entropy between every row of a large, sparse matrix
added 55 characters in body