User Matthew Drury

19 votes

Accepted

What is use of Tweedie or poisson loss/objective function in XGboost and Deep learning models

answered Oct 19, 2020 at 18:45

18 votes

Accepted

PCA finds a variable to be the most important twice

answered May 16, 2017 at 16:32

17 votes

Accepted

Is it wrong to choose features based on p-value?

answered Jul 12, 2017 at 19:44

17 votes

Accepted

How to handle predictors that are highly correlated to the response

answered Aug 5, 2016 at 15:18

17 votes

Probability of a single real-life future event: What does it mean when they say that "Hillary has a 75% chance of winning"?

answered Jul 24, 2016 at 16:07

17 votes

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

answered Oct 11, 2016 at 4:59

17 votes

What are some important uses of random number generation in computational statistics?

answered Jan 31, 2018 at 15:11

17 votes

Why does the variance of the Random walk increase?

answered Jul 2, 2015 at 17:49

17 votes

Are CDFs more fundamental than PDFs?

answered May 25, 2016 at 6:13

16 votes

Accepted

Is the sum of a large number of independent Cauchy random variables Normal?

answered May 20, 2016 at 18:40

16 votes

Accepted

Linear regression, conditional expectations and expected values

answered Jun 24, 2016 at 17:30

16 votes

Accepted

Choosing the range and grid density for regularization parameter in LASSO

answered Sep 30, 2015 at 20:14

15 votes

Is decision threshold a hyperparameter in logistic regression?

answered Jan 31, 2019 at 17:28

15 votes

Accepted

Is the median a "metric" or a "topological" property?

answered May 4, 2015 at 3:59

15 votes

Accepted

Interpretation of bs(x) and gam Results

answered May 4, 2015 at 4:26

15 votes

How to fit piecewise constant (or step-function) model and compare to logistic model in R

answered Jun 4, 2015 at 14:55

15 votes

Accepted

What is meant by the variance of functions in Introduction to Statistical Learning?

answered Apr 21, 2016 at 21:35

14 votes

Accepted

Do all bounded probability distributions have a definite mean?

answered Dec 26, 2019 at 17:30

14 votes

Why GLMs predict the mean and not the mode?

answered Sep 27, 2015 at 15:11

13 votes

Accepted

When is logit function preferred over sigmoid?

answered Nov 23, 2018 at 7:27

13 votes

Accepted

Is Multiple Linear Regression in 3 dimensions a plane of best fit or a line of best fit?

answered Jul 28, 2016 at 19:53

13 votes

Correlation between sine and cosine

answered Jul 16, 2016 at 21:34

13 votes

Accepted

Feature extracted by max pooling vs mean pooling

answered Jul 13, 2017 at 22:01

13 votes

Accepted

Overfitting with Linear Classifiers

answered Oct 6, 2015 at 22:28

13 votes

Accepted

Bias-variance decomposition and independence of $X$ and $\epsilon$

answered Aug 2, 2015 at 17:42

12 votes

Accepted

Is this really perfect separation in logistic regression, or is something else going on?

answered Nov 6, 2015 at 4:57

12 votes

Accepted

PCA: Eigenvectors of opposite sign and not being able to compute eigenvectors with `solve` in R

answered May 30, 2015 at 13:50

12 votes

Accepted

Why is the curse of dimensionality also called the empty space phenomenon?

answered May 5, 2015 at 22:08

12 votes

Accepted

Things that I am not sure about "LASSO" regression method

answered Jun 25, 2016 at 5:52

11 votes

Accepted

What to conclude when most results are statistically significant to fail to reject null hypothesis but not all?

answered Jun 4, 2017 at 22:55

Stack Exchange Network

432 Answers

What is use of Tweedie or poisson loss/objective function in XGboost and Deep learning models

PCA finds a variable to be the most important twice

Is it wrong to choose features based on p-value?

How to handle predictors that are highly correlated to the response

Probability of a single real-life future event: What does it mean when they say that "Hillary has a 75% chance of winning"?

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

What are some important uses of random number generation in computational statistics?

Why does the variance of the Random walk increase?

Are CDFs more fundamental than PDFs?

Is the sum of a large number of independent Cauchy random variables Normal?

Linear regression, conditional expectations and expected values

Choosing the range and grid density for regularization parameter in LASSO

Is decision threshold a hyperparameter in logistic regression?

Is the median a "metric" or a "topological" property?

Interpretation of bs(x) and gam Results

How to fit piecewise constant (or step-function) model and compare to logistic model in R

What is meant by the variance of functions in Introduction to Statistical Learning?

Do all bounded probability distributions have a definite mean?

Why GLMs predict the mean and not the mode?

When is logit function preferred over sigmoid?

Is Multiple Linear Regression in 3 dimensions a plane of best fit or a line of best fit?

Correlation between sine and cosine

Feature extracted by max pooling vs mean pooling

Overfitting with Linear Classifiers

Bias-variance decomposition and independence of $X$ and $\epsilon$

Is this really perfect separation in logistic regression, or is something else going on?

PCA: Eigenvectors of opposite sign and not being able to compute eigenvectors with `solve` in R

Why is the curse of dimensionality also called the empty space phenomenon?

Things that I am not sure about "LASSO" regression method

What to conclude when most results are statistically significant to fail to reject null hypothesis but not all?