User generic_user

152 votes

7 answers

122k views

Clustering on the output of t-SNE

asked Feb 23, 2017 at 1:39

130 votes

4 answers

130k views

When to use gamma GLMs?

asked Aug 16, 2013 at 8:13

48 votes

1 answer

16k views

Intuition behind tensor product interactions in GAMs (MGCV package in R)

asked Dec 8, 2012 at 21:30

42 votes

4 answers

17k views

Good methods for density plots of non-negative variables in R?

asked Jul 29, 2013 at 6:57

24 votes

7 answers

3k views

Explain in layperson's terms why predictive models aren't causally interpretable

asked Dec 1, 2021 at 14:03

22 votes

2 answers

17k views

Multiple imputation for outcome variables

asked Dec 19, 2012 at 9:03

14 votes

0 answers

454 views

Penalized spline confidence intervals based on cluster-sandwich VCV

asked Nov 30, 2012 at 19:39

14 votes

3 answers

4k views

Matrix form of backpropagation with batch normalization

asked Feb 12, 2018 at 18:57

12 votes

2 answers

12k views

Is Mahalanobis distance equivalent to the Euclidean one on the PCA-rotated data?

asked Aug 10, 2015 at 17:57

10 votes

0 answers

2k views

Dynamic Panel/GMM in R with group:time fixed effects? [closed]

asked Jul 10, 2014 at 23:33

9 votes

1 answer

3k views

Multivariate response regressions vs many linear models

asked Jun 7, 2013 at 5:47

8 votes

1 answer

1k views

Relative advantages of multiple imputation and expectation maximization (EM)

asked Feb 5, 2013 at 18:45

8 votes

1 answer

363 views

Causal identification and penalized splines

asked May 15, 2013 at 8:18

7 votes

1 answer

472 views

Maximum entropy distribution of a proportion with known mean and variance? Is it a beta?

asked Jul 25, 2016 at 17:41

5 votes

0 answers

771 views

Optimal block length for block bootstrap with multivariate time series

asked Apr 28, 2016 at 23:12

5 votes

0 answers

847 views

CDF of the ratio of two correlated $\chi^2$ random variables

asked Sep 24, 2015 at 19:00

5 votes

1 answer

300 views

causal inference with correlated multivariate outcomes

asked Mar 1, 2014 at 23:11

5 votes

1 answer

379 views

Equivalence of random effects via likelihood and smoothed splines

asked Sep 12, 2014 at 13:25

5 votes

0 answers

834 views

How general is the backfitting algorithm?

asked Jun 28, 2017 at 23:22

5 votes

1 answer

524 views

Sum or mean of across gradients in weight-sharing convolutional networks (and why)?

asked Sep 26, 2017 at 19:28

5 votes

1 answer

671 views

Mixed model by gradient descent?

asked Dec 6, 2017 at 22:32

5 votes

0 answers

522 views

BLUP formula for random coefficient models, in form that allows for convenient derivatives WRT loss function

asked Dec 6, 2017 at 23:40

4 votes

1 answer

231 views

matrix form of dummy expansion?

asked Oct 11, 2017 at 17:07

4 votes

0 answers

156 views

Optimal block size for spatial bootstrapping

asked Apr 23, 2018 at 18:57

4 votes

2 answers

355 views

Zeros in non-negative data: can you multiply by 10^c, round, and then model as a count variable?

asked Mar 15, 2014 at 0:33

4 votes

1 answer

2k views

Why don't my cluster-robust (panel-robust) standard errors match those in Stata? (solved)

asked Aug 7, 2013 at 11:21

4 votes

1 answer

182 views

Multiple imputation for variables used to calculate regression weights

asked May 27, 2013 at 12:57

4 votes

1 answer

204 views

Principled ways of constraining $E[Y]=0$ when one of your regressors $\rightarrow$ 0

asked Apr 15, 2013 at 0:47

4 votes

1 answer

374 views

Algorithm for selecting interactions without very many degrees of freedom (mgcv, gam)

asked Jan 18, 2013 at 17:24

4 votes

1 answer

2k views

Practical issues with dynamic panel data modeling

asked Jul 17, 2014 at 19:39

Stack Exchange Network

110 Questions

Clustering on the output of t-SNE

When to use gamma GLMs?

Intuition behind tensor product interactions in GAMs (MGCV package in R)

Good methods for density plots of non-negative variables in R?

Explain in layperson's terms why predictive models aren't causally interpretable

Multiple imputation for outcome variables

Penalized spline confidence intervals based on cluster-sandwich VCV

Matrix form of backpropagation with batch normalization

Is Mahalanobis distance equivalent to the Euclidean one on the PCA-rotated data?

Dynamic Panel/GMM in R with group:time fixed effects? [closed]

Multivariate response regressions vs many linear models

Relative advantages of multiple imputation and expectation maximization (EM)

Causal identification and penalized splines

Maximum entropy distribution of a proportion with known mean and variance? Is it a beta?

Optimal block length for block bootstrap with multivariate time series

CDF of the ratio of two correlated $\chi^2$ random variables

causal inference with correlated multivariate outcomes

Equivalence of random effects via likelihood and smoothed splines

How general is the backfitting algorithm?

Sum or mean of across gradients in weight-sharing convolutional networks (and why)?

Mixed model by gradient descent?

BLUP formula for random coefficient models, in form that allows for convenient derivatives WRT loss function

matrix form of dummy expansion?

Optimal block size for spatial bootstrapping

Zeros in non-negative data: can you multiply by 10^c, round, and then model as a count variable?

Why don't my cluster-robust (panel-robust) standard errors match those in Stata? (solved)

Multiple imputation for variables used to calculate regression weights

Principled ways of constraining $E[Y]=0$ when one of your regressors $\rightarrow$ 0

Algorithm for selecting interactions without very many degrees of freedom (mgcv, gam)

Practical issues with dynamic panel data modeling