User civilstat

31 votes

Accepted

Why do we need multivariate regression (as opposed to a bunch of univariate regressions)?

answered Jan 3, 2017 at 2:13

25 votes

Accepted

Should the y-axis on a survival plot go from 0 to 100 even if the lines are all above 0.9?

answered Mar 23, 2023 at 13:25

15 votes

What is/are the implicit priors in frequentist statistics?

answered Nov 25, 2016 at 20:07

15 votes

Why is the p-value defined the way it is (as opposed to a more intuitive measure)?

answered Nov 11, 2022 at 13:59

13 votes

Accepted

Can (some) linear regression model this (population) function accurately?

answered May 18, 2022 at 14:59

12 votes

Accepted

When sampling a population for surveys we can often limit our sample size to hundreds, but when doing a Monte Carlo simulation we need way more. Why?

answered Oct 23, 2022 at 3:54

11 votes

Use of splines in parameter estimation

answered Jun 19, 2023 at 19:38

11 votes

Why is lambda "within one standard error from the minimum" is a recommended value for lambda in an elastic net regression?

answered Nov 26, 2016 at 16:53

10 votes

Accepted

What is special about the Poisson/Binomial distributions such that they have special regression estimation techniques?

answered Apr 30 at 0:55

10 votes

Accepted

Is multicollinearity a "warning sign" for causal inference?

answered May 7 at 20:10

9 votes

Does k-fold cross validation strictly require shuffling of data before splitting it into k groups?

answered Oct 19, 2023 at 18:25

9 votes

Two worlds collide: Using ML for complex survey data

answered Oct 23, 2016 at 18:46

9 votes

Why are we using a biased and misleading standard deviation formula for $\sigma$ of a normal distribution?

answered Dec 11, 2016 at 2:37

9 votes

Why do we need alternative hypothesis?

answered Jan 14, 2019 at 0:46

8 votes

Odds "ratio" in logistic regression?

answered Apr 13, 2023 at 0:17

7 votes

How to motivate undergrads to take an Intro to Math Stats course

answered Feb 14, 2023 at 16:27

6 votes

$-z_\frac{\alpha}{2}$ and $z_\frac{\alpha}{2}$ in Confidence Intervals

answered Feb 19, 2023 at 13:45

6 votes

Are “Data are fixed” in Bayesian viewpoint and “Data are random” in frequentist viewpoint talking about the same thing mathematically?

answered Apr 10 at 2:21

6 votes

Accepted

What to do after Cross-Validation?

answered Oct 5, 2023 at 12:58

5 votes

Is it really so bad to do SMOTE on the training set before crossvalidation?

answered Mar 7 at 18:47

5 votes

Why divide data into 4 parts for IQR, and not into parts of 20 or 10 percentages each?

answered May 14 at 22:12

5 votes

Can I perform logistic regression or any other type of regression on this dataset?

answered Nov 29, 2023 at 19:45

5 votes

Accepted

How many times do I need to toss to differentiate between two biased coins?

answered Dec 7, 2023 at 14:15

5 votes

Stepwise regression predicts as well as Bayesian model averaging and boosting, why?

answered Nov 22, 2016 at 17:26

5 votes

Accepted

What is the best way to visualise a panel regression?

answered May 17, 2022 at 23:50

5 votes

Accepted

Should folds in k-fold CV actually be representative?

answered May 17, 2022 at 14:41

4 votes

How to calculate a permutation confidence interval?

answered Apr 29, 2022 at 20:34

4 votes

When are Shao's results on leave-one-out cross-validation applicable?

answered Nov 20, 2018 at 18:05

4 votes

Variance of $K$-fold cross-validation estimates as $f(K)$: what is the role of "stability"?

answered Jan 13, 2023 at 17:02

4 votes

Pick a card, any card: understanding the frequentist interpretation of a confidence interval

answered Sep 1, 2022 at 0:29

Stack Exchange Network

71 Answers

Why do we need multivariate regression (as opposed to a bunch of univariate regressions)?

Should the y-axis on a survival plot go from 0 to 100 even if the lines are all above 0.9?

What is/are the implicit priors in frequentist statistics?

Why is the p-value defined the way it is (as opposed to a more intuitive measure)?

Can (some) linear regression model this (population) function accurately?

When sampling a population for surveys we can often limit our sample size to hundreds, but when doing a Monte Carlo simulation we need way more. Why?

Use of splines in parameter estimation

Why is lambda "within one standard error from the minimum" is a recommended value for lambda in an elastic net regression?

What is special about the Poisson/Binomial distributions such that they have special regression estimation techniques?

Is multicollinearity a "warning sign" for causal inference?

Does k-fold cross validation strictly require shuffling of data before splitting it into k groups?

Two worlds collide: Using ML for complex survey data

Why are we using a biased and misleading standard deviation formula for $\sigma$ of a normal distribution?

Why do we need alternative hypothesis?

Odds "ratio" in logistic regression?

How to motivate undergrads to take an Intro to Math Stats course

$-z_\frac{\alpha}{2}$ and $z_\frac{\alpha}{2}$ in Confidence Intervals

Are “Data are fixed” in Bayesian viewpoint and “Data are random” in frequentist viewpoint talking about the same thing mathematically?

What to do after Cross-Validation?

Is it really so bad to do SMOTE on the training set before crossvalidation?

Why divide data into 4 parts for IQR, and not into parts of 20 or 10 percentages each?

Can I perform logistic regression or any other type of regression on this dataset?

How many times do I need to toss to differentiate between two biased coins?

Stepwise regression predicts as well as Bayesian model averaging and boosting, why?

What is the best way to visualise a panel regression?

Should folds in k-fold CV actually be representative?

How to calculate a permutation confidence interval?

When are Shao's results on leave-one-out cross-validation applicable?

Variance of $K$-fold cross-validation estimates as $f(K)$: what is the role of "stability"?

Pick a card, any card: understanding the frequentist interpretation of a confidence interval