Stack Exchange

Frank Harrell

biostat.mc.vanderbilt.edu/FrankHarrell

top accounts reputation activity favorites subscriptions
Top Questions
Top Answers
36
If mean is so sensitive, why use it in the first place?
stats.stackexchange.com

26
Given the power of computers these days, is there ever a reason to do a Chi-squared test rather than Fisher's exact test?
stats.stackexchange.com

21
Including the interaction but not the main effects in a model
stats.stackexchange.com

18
T-test for non normal when N>50?
stats.stackexchange.com

18
Can the MIC algorithm for detecting non-linear correlations be explained intuitively ?
stats.stackexchange.com

16
How to draw valid conclusions from “big data”?
stats.stackexchange.com

13
Should quantitative predictors be transformed to be normally distributed?
stats.stackexchange.com

12
Classic linear model - model selection
stats.stackexchange.com

10
Is there a good browser/viewer to see an R dataset (.rda file)
stats.stackexchange.com

10
What is your favorite layman's explanation for a difficult statistical concept?
stats.stackexchange.com

10
Choosing the best model from among different “best” models
stats.stackexchange.com

10
Classification vs. regression for prediction of the sign of a continuous response variable
stats.stackexchange.com

9
Is there any reason to prefer the AIC or BIC over the other?
stats.stackexchange.com

9
Reporting results of a logistic regression
stats.stackexchange.com

9
When to drop a term from a regression model?
stats.stackexchange.com

8
Minimum number of observations for logistic regression?
stats.stackexchange.com

8
Why continue to teach and use hypothesis testing?
stats.stackexchange.com

8
What are chunk tests?
stats.stackexchange.com

7
Model performance metrics for ordinal response
stats.stackexchange.com

7
Comparing importance of different sets of predictors
stats.stackexchange.com

7
What are good references containing arguments against null hypothesis significance testing?
stats.stackexchange.com

7
What do I gain if I consider the outcome as ordinal instead of categorical?
stats.stackexchange.com

7
Maximum number of independent variables that can be entered into a multiple regression equation
stats.stackexchange.com

7
Why is variable selection necessary?
stats.stackexchange.com

7
Do we need to report the median or the Mean when using a Kruskal-Wallis test?
stats.stackexchange.com

7
OLS Regression: Interaction effect is significant by itself, not when included in full model--ok to report?
stats.stackexchange.com

7
ANCOVA and its disturbing assumptions
stats.stackexchange.com

7
How do you decide what your train, validation and test percentages are?
stats.stackexchange.com

6
Is there an equivalent to Kruskal Wallis one-way test for a two-way model?
stats.stackexchange.com

6
Is it valid to include a baseline measure as control variable when testing the effect of an independent variable on change scores?
stats.stackexchange.com

1 2