User Jake Westfall

49 votes

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

answered Jul 10, 2020 at 4:40

41 votes

Maximum Likelihood Estimation (MLE) in layman terms

answered Aug 20, 2014 at 1:37

39 votes

Confused about independent probabilities. If a fair coin is flipped 5 times, P(HHHHH) = 0.03125, but P(H | HHHH) = 0.5

answered Jun 16, 2020 at 18:50

36 votes

Accepted

Big disagreement in the slope estimate when groups are treated as random vs. fixed in a mixed model

answered Aug 9, 2013 at 5:32

29 votes

Accepted

Why is the marginal distribution/marginal probability described as "marginal"?

answered May 15, 2019 at 1:48

27 votes

Accepted

Showing that 100 measurements for 5 subjects provide much less information than 5 measurements for 100 subjects

answered Jul 31, 2017 at 23:06

27 votes

Accepted

Why does treatment coding result in a correlation between random slope and intercept?

answered Apr 19, 2018 at 18:32

23 votes

Accepted

What is the upside of treating a factor as random in a mixed model?

answered Aug 28, 2014 at 21:18

22 votes

Suppression effect in regression: definition and visual explanation/depiction

answered Nov 2, 2013 at 1:00

21 votes

Why do I get zero variance of a random effect in my mixed model, despite some variation in the data?

answered Sep 11, 2014 at 19:28

21 votes

Accepted

What is the lme4::lmer equivalent of a three-way repeated measures ANOVA?

answered Nov 4, 2014 at 18:23

21 votes

Accepted

Why does statsmodels.api.OLS over-report the r-squared value?

answered Mar 14, 2017 at 7:42

20 votes

Accepted

Regression: why test normality of overall residuals, instead of residuals conditional on $\hat{y}$?

answered Apr 24, 2016 at 1:07

20 votes

Accepted

Intraclass Correlation Coefficient in mixed model with random slopes

answered Dec 12, 2017 at 14:25

19 votes

Accepted

How to partition the variance explained at group level and at individual level?

answered Aug 14, 2013 at 1:08

19 votes

Accepted

In math/stats terminology, $1/p$ is the reciprocal of $p$. So what is $1-p$ called, if anything?

answered Jun 5, 2013 at 20:32

17 votes

Accepted

Equivalence between a repeated measures anova model and a mixed model: lmer vs lme, and compound symmetry

answered Nov 14, 2014 at 0:59

16 votes

z-score and Normal Distribution

answered Sep 1, 2017 at 15:42

16 votes

Accepted

What does pooled variance "actually" mean?

answered Sep 13, 2017 at 2:39

16 votes

Accepted

Is R's glm function useless in a big data / machine learning setting?

answered Oct 11, 2016 at 5:00

16 votes

Is up- or down-sampling imbalanced data actually that effective? Why?

answered Jun 28, 2020 at 3:01

15 votes

Accepted

Mixed effects model: Compare random variance component across levels of a grouping variable

answered Aug 19, 2018 at 3:13

15 votes

Does $r$-squared have a $p$-value?

answered Aug 12, 2014 at 21:00

15 votes

Adding random effect influences coefficient estimates

answered Oct 23, 2013 at 0:51

14 votes

Relationships between correlation and causation

answered Jun 2, 2017 at 15:49

12 votes

Accepted

Is there a rule of thumb for using standard deviation of k-fold cross-validation scores to pick the best model?

answered Aug 12, 2016 at 1:48

12 votes

Accepted

Regressing out a variable?

answered Sep 21, 2014 at 22:38

12 votes

Accepted

Can I probe cross-level interactions without random slope in a hierarchical linear model?

answered May 23, 2013 at 19:13

12 votes

Accepted

Equivalent to Welch's t-test in GLS framework

answered Apr 2, 2015 at 1:59

11 votes

Accepted

Multilevel model with responses only at level 2

answered Sep 1, 2015 at 2:42

Stack Exchange Network

116 Answers

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

Maximum Likelihood Estimation (MLE) in layman terms

Confused about independent probabilities. If a fair coin is flipped 5 times, P(HHHHH) = 0.03125, but P(H | HHHH) = 0.5

Big disagreement in the slope estimate when groups are treated as random vs. fixed in a mixed model

Why is the marginal distribution/marginal probability described as "marginal"?

Showing that 100 measurements for 5 subjects provide much less information than 5 measurements for 100 subjects

Why does treatment coding result in a correlation between random slope and intercept?

What is the upside of treating a factor as random in a mixed model?

Suppression effect in regression: definition and visual explanation/depiction

Why do I get zero variance of a random effect in my mixed model, despite some variation in the data?

What is the lme4::lmer equivalent of a three-way repeated measures ANOVA?

Why does statsmodels.api.OLS over-report the r-squared value?

Regression: why test normality of overall residuals, instead of residuals conditional on $\hat{y}$?

Intraclass Correlation Coefficient in mixed model with random slopes

How to partition the variance explained at group level and at individual level?

In math/stats terminology, $1/p$ is the reciprocal of $p$. So what is $1-p$ called, if anything?

Equivalence between a repeated measures anova model and a mixed model: lmer vs lme, and compound symmetry

z-score and Normal Distribution

What does pooled variance "actually" mean?

Is R's glm function useless in a big data / machine learning setting?

Is up- or down-sampling imbalanced data actually that effective? Why?

Mixed effects model: Compare random variance component across levels of a grouping variable

Does $r$-squared have a $p$-value?

Adding random effect influences coefficient estimates

Relationships between correlation and causation

Is there a rule of thumb for using standard deviation of k-fold cross-validation scores to pick the best model?

Regressing out a variable?

Can I probe cross-level interactions without random slope in a hierarchical linear model?

Equivalent to Welch's t-test in GLS framework

Multilevel model with responses only at level 2