21,161 reputation
12979
bio website biostat.mc.vanderbilt.edu/…
location Nashville, TN
age
visits member for 3 years, 6 months
seen 1 hour ago

I am Professor of Biostatistics and Chairman of the Department of Biostatistics at Vanderbilt University School of Medicine, Nashville TN USA. I am Associate Editor of Statistics in Medicine, a member of the Faculty of 1000 Medicine, and a member of the policy advisory board for the Journal of Clinical Epidemiology. I am a Fellow of the American Statistical Association. I am author of Regression Modeling Strategies (Springer, 2001). My specialties are development and validation of predictive models, clinical trials, observational clinical research, cardiovascular research, technology evaluation, clinical epidemiology, medical diagnostic accuracy, biomarker research, pharmaceutical safety, Bayesian methods, quantifying predictive accuracy, missing data imputation, and statistical graphics and reporting. I am a long-time user of R. In August 2014 I was given the WJ Dixon Award for Excellence in Statistical Consulting by the American Statistical Association. Among many other things, Dr Dixon was the lead developer of the first general-purpose statistical software package, BMD.


13h
comment Why square the difference instead of taking the absolute value in standard deviation?
I remain unconvinced that variances are very useful for asymmetric distributions.
1d
comment Range of predicted probabilities by logistic regression
Is this really a classification problem? Or is it the task of estimating risk? For the latter you may not need a threshold. In general a threshold is needed only if a decision is needed automatically and instantaneously, and utilities are not considered.
1d
comment Generate predictions from a logistic regression model reflecting the uncertainty of the model
If you still want a sampling approach you can use the "poor man's Bayesian approach" of bootstrapping whereby you develop (from scratch) several models from samples with replacement from the original dataset and show the distributions of predicted values. The 0.025 and 0.975 quantiles of predictions for one subject would provide a 0.95 confidence interval for the unknown true probability of an event for that subject, which is the long way to obtain a confidence interval, but sometimes more accurate than Gaussian theory-based Wald intervals.
Oct
19
comment weibull mle with optim in R
The second sentence is the sort of response I was hoping for. I hope that someone can help you with this. My only thought is to see if reparametizing as survreg does will help.
Oct
19
comment weibull mle with optim in R
Please let us know why you are not using the survreg function in the R survival package.
Oct
19
revised Is the convention for error bars to present one or two standard errors?
Added a how-to for raw data instead of point estimate
Oct
19
answered Is the convention for error bars to present one or two standard errors?
Oct
18
comment Is it acceptable to transform data for use in a GLM using Poisson?
There are so many resources for linear models that it's hard to know where to begin. I hope that others will respond with book recommendations.
Oct
18
revised Is it acceptable to transform data for use in a GLM using Poisson?
edited tags
Oct
18
answered Is it acceptable to transform data for use in a GLM using Poisson?
Oct
18
comment Linear regresson lm or stepwise regression here using R?
To add to Peter's comment and answer, regression models don't care about the distribution of $X$s. In the linear model one only cares about the distribution of residuals from the model. And there is no principle in statistics that would suggest that simply using the fully pre-specified model is not a good approach, once you understand co-linearities and perhaps get "chunk" tests that combine competing variables into a multiple degree of freedom test.
Oct
17
comment When can a continuous variable be treated as categorical?
Well said. I wish I had worded it in a way to make clear that I wasn't criticizing the OP but rather the common practice.
Oct
17
answered When can a continuous variable be treated as categorical?
Oct
17
revised What's wrong with Bonferroni adjustments
elaborated on inconsistent thinking coming from multiplicity adjustments
Oct
17
comment What's wrong with Bonferroni adjustments
To answer @MJA I think there are two preferred approaches: (1) be Bayesian or (2) prioritize the hypotheses and report the results in context, in priority order.
Oct
17
comment What's wrong with Bonferroni adjustments
Not clear on the reference to 'repeatable'. If there is a single test, with no multiplicity adjustment required, the chance that a result with $P=0.04$ is repeated is not high.
Oct
17
answered What's wrong with Bonferroni adjustments
Oct
14
revised How to interpret variables that are excluded from or included in the lasso model?
Added text for the binary Y case
Oct
14
comment How to interpret variables that are excluded from or included in the lasso model?
Last comment. I clarified this in my long answer above.
Oct
14
comment How to interpret variables that are excluded from or included in the lasso model?
I need to sign off this discussion - the basic answer to your question is basic R programming plus take a look at simple simulations in biostat.mc.vanderbilt.edu/rms.