User 24n8

8 votes

3 answers

1k views

Why can $R^2$ be negative in linear regression -- interview question [duplicate]

asked Aug 7, 2020 at 2:54

6 votes

1 answer

393 views

Is the regressor (sometimes called "independent" variable) actually independent of the response from a probabilistic perspective?

asked Jul 23, 2020 at 21:30

6 votes

1 answer

153 views

Confidence interval example in Computer Age Statistical Inference

asked Jul 8, 2020 at 16:50

5 votes

1 answer

709 views

Supplemental material for Elements of Statistical Learning?

asked May 18, 2020 at 0:23

5 votes

1 answer

537 views

Is stochastic gradient descent pseudo-stochastic?

asked Mar 20, 2019 at 15:14

5 votes

1 answer

1k views

What is the purpose of using duplicated data in resampling techniques (e.g., bagging/bootstrapping)?

asked Sep 3, 2020 at 18:09

5 votes

3 answers

4k views

How do you learn labels with unsupervised learning?

asked Aug 30, 2021 at 3:20

5 votes

1 answer

62 views

When does adding a new predictor not increase $R^2$ in OLS?

asked Nov 24 at 22:10

4 votes

2 answers

862 views

Are regression trees actually performing "regression?"

asked Sep 20, 2020 at 16:13

4 votes

1 answer

57 views

Uniform distribution presentations seem incomplete

asked May 2, 2020 at 13:34

3 votes

2 answers

810 views

Neural networks for regression vs. more classical regression methods?

asked Dec 31, 2018 at 19:20

3 votes

2 answers

3k views

Multivariate regression vs. multiple univariate regression models

asked Dec 31, 2018 at 20:03

3 votes

1 answer

98 views

What does $cov(x_1,x_2) >> 0, cov(y_1, y_2) >> 0$ and $cov(x_1+y_1, x_2+y_2) = 0$ tell us about $x_1, x_2, y_1, y_2$?

asked May 6, 2020 at 1:36

3 votes

0 answers

79 views

In practice, do we distinguish between "in-sample" and "training" error?

asked Aug 20, 2020 at 17:42

3 votes

1 answer

298 views

Is there theoretical guarantee that principal components are orthogonal?

asked May 6, 2021 at 16:13

3 votes

1 answer

143 views

Outliers for classification; imbalanced data for regression?

asked Aug 30, 2021 at 2:39

2 votes

1 answer

55 views

Why are linear regression models used for things like pricing or modeling a response that can only be positive?

asked May 10, 2021 at 0:36

2 votes

1 answer

298 views

Why does centering/standardizing the regressor make the p value for the intercept zero?

asked May 4, 2021 at 14:13

2 votes

1 answer

1k views

Duplicating a dataset for OLS and effects on variance

asked Aug 24, 2020 at 22:23

2 votes

0 answers

29 views

Under which scenarios would $R^2$ increase the most when adding a new predictor (linear independent, uncorrelated, orthogonal)?

asked Aug 12, 2020 at 3:26

2 votes

1 answer

108 views

Why can 'linear' regressions include non-linear transformations of the independent variables?

asked Jun 25, 2020 at 18:44

1 vote

0 answers

29 views

Given $N$ IID zero-meaned points on a plane, find the equation for the plane

asked Jul 1, 2020 at 4:16

1 vote

0 answers

36 views

Multicollinearity negating $\beta$

asked Jul 10, 2020 at 4:05

1 vote

0 answers

113 views

Regularization and multicollinearity

asked May 9, 2020 at 15:19

1 vote

2 answers

186 views

Why do sources claim that linear regression assumes little to no multicollinearity, among other "assumptions"?

asked May 22, 2020 at 4:21

1 vote

1 answer

66 views

Why do we use the term "uncorrelated" to describe linear dependence/independence?

asked May 3, 2020 at 15:01

1 vote

1 answer

88 views

Randomly choosing objects from a set of objects - guarantee all objects are picked at least once?

asked Mar 31, 2019 at 2:29

1 vote

2 answers

932 views

What is the difference between having input parameters that are independent/dependent in ML tasks?

asked Apr 24, 2019 at 15:25

1 vote

1 answer

51 views

What's a good resource for learning about applying linear regression to practical problems?

asked Sep 4, 2020 at 15:56

1 vote

1 answer

19 views

Is there formal terminology for specific phenomena that happens in multiple regression?

asked Apr 21, 2021 at 18:50

Stack Exchange Network

36 Questions

Why can $R^2$ be negative in linear regression -- interview question [duplicate]

Is the regressor (sometimes called "independent" variable) actually independent of the response from a probabilistic perspective?

Confidence interval example in Computer Age Statistical Inference

Supplemental material for Elements of Statistical Learning?

Is stochastic gradient descent pseudo-stochastic?

What is the purpose of using duplicated data in resampling techniques (e.g., bagging/bootstrapping)?

How do you learn labels with unsupervised learning?

When does adding a new predictor not increase $R^2$ in OLS?

Are regression trees actually performing "regression?"

Uniform distribution presentations seem incomplete

Neural networks for regression vs. more classical regression methods?

Multivariate regression vs. multiple univariate regression models

What does $cov(x_1,x_2) >> 0, cov(y_1, y_2) >> 0$ and $cov(x_1+y_1, x_2+y_2) = 0$ tell us about $x_1, x_2, y_1, y_2$?

In practice, do we distinguish between "in-sample" and "training" error?

Is there theoretical guarantee that principal components are orthogonal?

Outliers for classification; imbalanced data for regression?

Why are linear regression models used for things like pricing or modeling a response that can only be positive?

Why does centering/standardizing the regressor make the p value for the intercept zero?

Duplicating a dataset for OLS and effects on variance

Under which scenarios would $R^2$ increase the most when adding a new predictor (linear independent, uncorrelated, orthogonal)?

Why can 'linear' regressions include non-linear transformations of the independent variables?

Given $N$ IID zero-meaned points on a plane, find the equation for the plane

Multicollinearity negating $\beta$

Regularization and multicollinearity

Why do sources claim that linear regression assumes little to no multicollinearity, among other "assumptions"?

Why do we use the term "uncorrelated" to describe linear dependence/independence?

Randomly choosing objects from a set of objects - guarantee all objects are picked at least once?

What is the difference between having input parameters that are independent/dependent in ML tasks?

What's a good resource for learning about applying linear regression to practical problems?

Is there formal terminology for specific phenomena that happens in multiple regression?