User EdM

58 votes

Accepted

Is it true that the percentile bootstrap should never be used?

answered Jul 17, 2018 at 2:02

42 votes

Accepted

Why is the logistic distribution called "logistic"?

answered Jan 28, 2016 at 22:09

37 votes

Accepted

Is it possible to interpret the bootstrap from a Bayesian perspective?

answered Oct 3, 2013 at 19:13

37 votes

How much lung cancer is really caused by smoking?

answered Jun 8, 2018 at 19:24

35 votes

Accepted

In Regression Analysis, why do we call independent variables "independent"?

answered Jul 18, 2018 at 12:03

33 votes

Accepted

predict() Function for lmer Mixed Effects Models

answered Sep 25, 2015 at 23:07

30 votes

Accepted

Ridge\Lasso -- Standardization of dummy indicators

answered Jul 27, 2018 at 14:49

27 votes

Accepted

The order of variables in ANOVA matters, doesn't it?

answered May 18, 2016 at 21:29

26 votes

Fitting known equation to data

answered Sep 4, 2015 at 23:52

22 votes

Accepted

Ridge regression in R with p values and goodness of fit

answered Sep 7, 2015 at 16:05

20 votes

Are all models useless? Is any exact model possible -- or useful?

answered May 3, 2016 at 22:50

19 votes

How to obtain Confidence Intervals for a LASSO regression?

answered Apr 10, 2019 at 16:27

19 votes

Accepted

Why lasso for feature selection?

answered Sep 16, 2018 at 16:21

19 votes

Accepted

Bootstrap methodology. Why resample "with replacement" instead of random subsampling?

answered Sep 7, 2015 at 14:41

18 votes

Accepted

LASSO with interaction terms - is it okay if main effects are shrunk to zero?

answered Nov 14, 2016 at 21:21

18 votes

Accepted

Calculating F-Score, which is the "positive" class, the majority or minority class?

answered Jan 27, 2016 at 16:49

17 votes

Accepted

How much missing data is too much? Multiple Imputation (MICE) & R

answered May 1, 2015 at 16:23

17 votes

For linear classifiers, do larger coefficients imply more important features?

answered Mar 21, 2016 at 18:57

16 votes

coxph ran out of iterations and did not converge

answered Aug 6, 2013 at 17:18

16 votes

log mean vs mean log in statistics

answered Dec 7, 2016 at 15:49

16 votes

Accepted

When we plot data and then use nonlinear transformations in a regression model are we data-snooping?

answered Aug 22, 2020 at 16:23

16 votes

Accepted

Why is it OK to model demographics as random effects in bayesian multilevel models?

answered Jul 15, 2020 at 16:59

15 votes

Accepted

What are the use cases for Propensity Score Matching?

answered Apr 30, 2021 at 12:21

15 votes

Calculating p values for data that is less than 1

answered Dec 1, 2016 at 16:41

15 votes

Accepted

Why do p-values change in significance when changing the order of covariates in the aov model?

answered May 15, 2016 at 18:31

15 votes

What are the worst (commonly adopted) ideas/principles in statistics?

answered Jul 10, 2020 at 12:47

15 votes

Accepted

I've already used my entire dataset in a regression, should I not use that as a prediction model?

answered Oct 25, 2021 at 19:46

13 votes

How to handle NA values in shrinkage (Lasso) method using glmnet

answered May 13, 2015 at 17:58

13 votes

Biased bootstrap: is it okay to center the CI around the observed statistic?

answered Jun 25, 2015 at 18:49

13 votes

Accepted

Family-wise error boundary: Does re-using data sets on different studies of independent questions lead to multiple testing problems?

answered Aug 22, 2015 at 21:10

Stack Exchange Network

3,032 Answers

Is it true that the percentile bootstrap should never be used?

Why is the logistic distribution called "logistic"?

Is it possible to interpret the bootstrap from a Bayesian perspective?

How much lung cancer is really caused by smoking?

In Regression Analysis, why do we call independent variables "independent"?

predict() Function for lmer Mixed Effects Models

Ridge\Lasso -- Standardization of dummy indicators

The order of variables in ANOVA matters, doesn't it?

Fitting known equation to data

Ridge regression in R with p values and goodness of fit

Are all models useless? Is any exact model possible -- or useful?

How to obtain Confidence Intervals for a LASSO regression?

Why lasso for feature selection?

Bootstrap methodology. Why resample "with replacement" instead of random subsampling?

LASSO with interaction terms - is it okay if main effects are shrunk to zero?

Calculating F-Score, which is the "positive" class, the majority or minority class?

How much missing data is too much? Multiple Imputation (MICE) & R

For linear classifiers, do larger coefficients imply more important features?

coxph ran out of iterations and did not converge

log mean vs mean log in statistics

When we plot data and then use nonlinear transformations in a regression model are we data-snooping?

Why is it OK to model demographics as random effects in bayesian multilevel models?

What are the use cases for Propensity Score Matching?

Calculating p values for data that is less than 1

Why do p-values change in significance when changing the order of covariates in the aov model?

What are the worst (commonly adopted) ideas/principles in statistics?

I've already used my entire dataset in a regression, should I not use that as a prediction model?

How to handle NA values in shrinkage (Lasso) method using glmnet

Biased bootstrap: is it okay to center the CI around the observed statistic?

Family-wise error boundary: Does re-using data sets on different studies of independent questions lead to multiple testing problems?