User Lepidopterist

17 votes

2 answers

3k views

High variance of the distribution of p-values (an argument in Taleb 2016)

asked Jun 27, 2016 at 19:17

12 votes

1 answer

3k views

Efficient/fast Mahalanobis distance computation

asked Apr 19, 2015 at 18:19

10 votes

1 answer

2k views

What is the intuitive (geometric?) meaning of minimizing the log determinant of a matrix?

asked May 7, 2015 at 20:39

10 votes

1 answer

3k views

Order statistic for beta distribution

asked Jun 12, 2015 at 18:12

7 votes

1 answer

2k views

Cholesky factorization and forward substitution less accurate than inversion?

asked Apr 22, 2015 at 1:00

7 votes

1 answer

567 views

Is there a counterexample to the claim that throwing away "insignificant" predictors doesn't generally harm a model?

asked Mar 20, 2017 at 15:47

6 votes

2 answers

926 views

Distribution of "sample" mahalanobis distances

asked Jun 13, 2015 at 19:44

6 votes

1 answer

1k views

Asymptotic distribution of $\chi^2_{(1)}$

asked Feb 3, 2015 at 0:01

5 votes

1 answer

570 views

What is a two-stage regression, as a prelude to multilevel modeling, concretely?

asked Nov 27, 2018 at 15:18

4 votes

0 answers

227 views

Can I get away with using GLM models on "pseudo-panel" big N small time=T data?

asked Mar 27, 2017 at 17:12

4 votes

1 answer

55 views

Is there a convincing case to be made against using last years data as a holdout set when building a predictive model?

asked Feb 27, 2017 at 22:29

4 votes

0 answers

103 views

Outlier removal using a notion of statistical distance?

asked Jan 30, 2017 at 21:35

3 votes

0 answers

109 views

Does a "pruned" i.i.d. multivariate sample behave as the i.i.d. sample?

asked Jun 17, 2015 at 1:34

3 votes

1 answer

4k views

Bayes estimator under LINEX loss?

asked Nov 3, 2014 at 3:39

3 votes

0 answers

2k views

Fast Kendall Tau in Matlab (and if necessary another language)

asked Jun 29, 2014 at 16:01

3 votes

0 answers

269 views

On the expectation of the reciprocal of a quadratic form of a standard normal variable

asked Dec 15, 2015 at 17:46

3 votes

1 answer

2k views

Can one usefully specify a multilevel-model with a partially-nested, partially non-nested structure?

asked Jan 7, 2019 at 22:31

2 votes

0 answers

19 views

Understanding how sensitive a model is to a predictor only implicitly part of the model

asked Mar 20, 2017 at 1:27

2 votes

0 answers

412 views

Elastic net for general GLMs (in particular the Gamma family)

asked Jun 9, 2017 at 14:25

2 votes

1 answer

3k views

nearest neighbors degrees of freedom

asked Feb 20, 2013 at 4:35

2 votes

1 answer

80 views

Testing whether we should build a model on the union of data sets

asked Feb 21, 2017 at 18:54

1 vote

0 answers

34 views

Tailbounds for a rational function of the average of chi-squares

asked Dec 13, 2015 at 20:31

1 vote

0 answers

24 views

Distribution of the maximum vector (over i.i.d. set) and its dot product with the eigenvectors spanning the rest.

asked Jun 12, 2015 at 17:42

1 vote

1 answer

595 views

Reconstructing a vector after projection

asked Jul 9, 2014 at 14:54

1 vote

1 answer

63 views

Seeking regression modeling strategies for predicting prices based on categorical variables (one of which is ordered)

asked Nov 11, 2017 at 22:09

1 vote

0 answers

161 views

Am I doing too many likelihood ratio tests to simplify the model?

asked Mar 1, 2017 at 15:15

0 votes

0 answers

79 views

Learning distance metric from output of knn

asked Jul 6, 2019 at 17:31

0 votes

0 answers

225 views

Interpreting the inverse covariance matrix: $S^{-1}x$ and $x^T S^{-1}x$ [duplicate]

asked Oct 3, 2014 at 20:06

0 votes

0 answers

115 views

Probability that the entries of a sample covariance matrix of a m.v. Gaussian exceeds a threshold

asked Oct 21, 2014 at 0:22

0 votes

0 answers

124 views

Parametrizing a matrix (and algorithm) by its orthogonal complement

asked Oct 28, 2014 at 19:05

Stack Exchange Network

31 Questions

High variance of the distribution of p-values (an argument in Taleb 2016)

Efficient/fast Mahalanobis distance computation

What is the intuitive (geometric?) meaning of minimizing the log determinant of a matrix?

Order statistic for beta distribution

Cholesky factorization and forward substitution less accurate than inversion?

Is there a counterexample to the claim that throwing away "insignificant" predictors doesn't generally harm a model?

Distribution of "sample" mahalanobis distances

Asymptotic distribution of $\chi^2_{(1)}$

What is a two-stage regression, as a prelude to multilevel modeling, concretely?

Can I get away with using GLM models on "pseudo-panel" big N small time=T data?

Is there a convincing case to be made against using last years data as a holdout set when building a predictive model?

Outlier removal using a notion of statistical distance?

Does a "pruned" i.i.d. multivariate sample behave as the i.i.d. sample?

Bayes estimator under LINEX loss?

Fast Kendall Tau in Matlab (and if necessary another language)

On the expectation of the reciprocal of a quadratic form of a standard normal variable

Can one usefully specify a multilevel-model with a partially-nested, partially non-nested structure?

Understanding how sensitive a model is to a predictor only implicitly part of the model

Elastic net for general GLMs (in particular the Gamma family)

nearest neighbors degrees of freedom

Testing whether we should build a model on the union of data sets

Tailbounds for a rational function of the average of chi-squares

Distribution of the maximum vector (over i.i.d. set) and its dot product with the eigenvectors spanning the rest.

Reconstructing a vector after projection

Seeking regression modeling strategies for predicting prices based on categorical variables (one of which is ordered)

Am I doing too many likelihood ratio tests to simplify the model?

Learning distance metric from output of knn

Interpreting the inverse covariance matrix: $S^{-1}x$ and $x^T S^{-1}x$ [duplicate]

Probability that the entries of a sample covariance matrix of a m.v. Gaussian exceeds a threshold

Parametrizing a matrix (and algorithm) by its orthogonal complement