39,555 reputation
53490
bio website statisticalanalysisconsulting…
location New York, NY
age 54
visits member for 3 years, 8 months
seen 8 hours ago

I'm a statistical consultant to graduate students and researchers in fields including the behavioral and health sciences. I've assisted with review of articles, and with preparation of grants, dissertations and papers.


14h
comment Regresssion of Accurate Data
Do you mean that you have very little data, but it is very accurate?
15h
comment Regresssion of Accurate Data
Then I don't understand what you have or what you want.
17h
comment Regresssion of Accurate Data
First you say you have perfectly accurate data. Then you say there are imperfections in the measuring device. I'm confused.
17h
reviewed No Action Needed Regresssion of Accurate Data
23h
answered Repeated measures tests without “before and after” measurements
23h
comment Cox PH model selection and validation
No, don't use backward stepwise at all. Start with your a priori model and add variables that change the parameters in important ways or that improve fit. That's your model. It may be that there are several that are good - you can compare them with AIC or BIC
23h
reviewed Close Difference in memory usage between gbm and blackboost
23h
reviewed Leave Open Cluster with distance threshold in R
23h
reviewed Close How to interpret results from a nonparametric Wilcoxon test
23h
reviewed Leave Open Feature Selection for look alike modeling using k-NN
23h
answered Cox PH model selection and validation
1d
answered Forecasting discrete non normal data
1d
comment Interaction in Survival analysis
When there is a large interaction effect, the analysis of main effects alone is not appropriate. You can get all kinds of results.
1d
answered Interaction in Survival analysis
2d
comment Interaction in Survival analysis
Did you include the main effect of Lot1 in the final analysis you show? You should.
2d
comment How to compare two different clustering approaches?
In addition to statistical methods, I would urge you to look at the substantive meaning: Which clustering is more useful to you?
2d
comment Considering correlation of independent variables in regression model
As far as I know, you have to compare the models - that's what any package would do, anyway
2d
comment Meaning of intercept in multinomial regression with binary predictors?
That would be cool
Apr
15
reviewed Close How to merge datasets in STATA conditionally?
Apr
15
comment How to test stationary for variables in ARDL model
This question appears to be off-topic because it is about how to do something in Stata. It would be better asked on the Stata list