User jwimberley

154 votes

Why is Newton's method not widely used in machine learning?

answered Dec 29, 2016 at 1:19

18 votes

Accepted

What is the name of the distribution with a probability density like $1/(1+\exp(x))$?

answered Dec 15, 2016 at 1:06

16 votes

Accepted

If I have the expected value of the logarithm of a RV, can I obtain the expected value of the RV itself?

answered Sep 19, 2016 at 15:42

15 votes

Accepted

Linear model where the data has uncertainty, using R

answered Sep 19, 2016 at 12:15

12 votes

Is there a formal name for this statistical fallacy?

answered Sep 30, 2021 at 14:32

11 votes

Null model AIC lower than full model AIC in logistic regression in R

answered Feb 27, 2017 at 22:06

7 votes

Accepted

Generating random numbers that are log-normally distributed

answered Aug 5, 2020 at 14:28

5 votes

What does the Akaike Information Criterion (AIC) score of a model mean?

answered Mar 21, 2017 at 13:48

5 votes

Can we really sample from a Continuous distribution (Scipy function) and what does it mean?

answered Sep 19, 2016 at 1:50

5 votes

Accepted

Integral identity of lemma contained in infoGAN paper

answered Dec 19, 2016 at 17:09

5 votes

Accepted

One-vs-All Logistic Regression Probability Scores either close to 0 or 1

answered Dec 30, 2016 at 19:05

5 votes

How versatile is logistic regression?

answered Jan 6, 2017 at 22:24

4 votes

Accepted

Confidence interval for first order linear regression

answered Jan 17, 2017 at 14:22

4 votes

Is 50% 100% higher than 25% or is it 25% higher than 25%?

answered Feb 16, 2017 at 15:47

4 votes

Accepted

Recommended Mathematics electives for Machine Learning

answered Feb 25, 2017 at 0:06

4 votes

Why are hypothesis tests still used when we have the bootstrap and central limit theorem?

answered Oct 23, 2021 at 14:29

4 votes

$\mathbb{E} \left[ \frac{X}{c + X} \right]$ for known constant $c > 0$?

answered Feb 25, 2022 at 2:41

3 votes

Accepted

What can go wrong with MLE if I substitute some first-stage estimates instead of some parameters?

answered Jan 31, 2017 at 16:45

3 votes

Prediction plot from binomial glm/regression - prediction line does not go up to 1 on the y-axis

answered Sep 28, 2021 at 14:02

3 votes

Random variables for which Markov, Chebyshev inequalities are tight

answered Sep 18, 2016 at 1:12

2 votes

Regression on the unit disk starting from "uniformly spaced" samples

answered Sep 18, 2016 at 20:18

2 votes

Test to determine if a set of coins have specified biases

answered Sep 21, 2016 at 17:59

2 votes

Similarity between time-dependent paths

answered Sep 23, 2016 at 13:09

2 votes

Indicator variable for binary data: {-1,1} vs {0,1}

answered Nov 7, 2016 at 13:35

2 votes

Relationship between binomial regression link function and goodness-of-fit tests [now with link to R code]

answered Nov 23, 2016 at 14:27

2 votes

Does it make sense to think about a distribution of distributions?

answered Dec 11, 2016 at 15:33

2 votes

Accepted

Linear regression of B-splines with terms inside an integral?

answered Jan 23, 2017 at 18:46

2 votes

Testing difference in (dependent) slopes from a multiple regression?

answered Dec 20, 2016 at 16:15

2 votes

n random variables' sum equals 1 , and they belong to the same distribution and are restricted in [0,1] , what distribution do they obey?

answered Dec 18, 2016 at 15:03

2 votes

Distribution of the product of two LogNormals

answered Apr 18, 2022 at 13:35

Stack Exchange Network

59 Answers

Why is Newton's method not widely used in machine learning?

What is the name of the distribution with a probability density like $1/(1+\exp(x))$?

If I have the expected value of the logarithm of a RV, can I obtain the expected value of the RV itself?

Linear model where the data has uncertainty, using R

Is there a formal name for this statistical fallacy?

Null model AIC lower than full model AIC in logistic regression in R

Generating random numbers that are log-normally distributed

What does the Akaike Information Criterion (AIC) score of a model mean?

Can we really sample from a Continuous distribution (Scipy function) and what does it mean?

Integral identity of lemma contained in infoGAN paper

One-vs-All Logistic Regression Probability Scores either close to 0 or 1

How versatile is logistic regression?

Confidence interval for first order linear regression

Is 50% 100% higher than 25% or is it 25% higher than 25%?

Recommended Mathematics electives for Machine Learning

Why are hypothesis tests still used when we have the bootstrap and central limit theorem?

$\mathbb{E} \left[ \frac{X}{c + X} \right]$ for known constant $c > 0$?

What can go wrong with MLE if I substitute some first-stage estimates instead of some parameters?

Prediction plot from binomial glm/regression - prediction line does not go up to 1 on the y-axis

Random variables for which Markov, Chebyshev inequalities are tight

Regression on the unit disk starting from "uniformly spaced" samples

Test to determine if a set of coins have specified biases

Similarity between time-dependent paths

Indicator variable for binary data: {-1,1} vs {0,1}

Relationship between binomial regression link function and goodness-of-fit tests [now with link to R code]

Does it make sense to think about a distribution of distributions?

Linear regression of B-splines with terms inside an integral?

Testing difference in (dependent) slopes from a multiple regression?

n random variables' sum equals 1 , and they belong to the same distribution and are restricted in [0,1] , what distribution do they obey?

Distribution of the product of two LogNormals