User andrewH

27 votes

3 answers

2k views

Having a conjugate prior: Deep property or mathematical accident?

asked May 18, 2013 at 6:42

18 votes

5 answers

14k views

Does independence imply conditional independence?

asked Mar 5, 2013 at 0:04

16 votes

2 answers

1k views

What if there is no true data-generating process?

asked Sep 3, 2021 at 2:53

13 votes

1 answer

4k views

Constructing a continuous distribution to match $m$ moments

asked Mar 13, 2015 at 18:27

10 votes

1 answer

1k views

Penalized methods for categorical data: combining levels in a factor

asked May 27, 2013 at 2:36

9 votes

1 answer

2k views

Does every log-linear model have a perfectly equivalent logistic regression?

asked Mar 27, 2013 at 22:45

9 votes

1 answer

6k views

Can I get the parameters of a lognormal distribution from the sample mean & median?

asked Sep 28, 2015 at 2:03

6 votes

2 answers

431 views

Getting the units right for the Pareto distribution of wealth: x = people, dollars, dollars per person?

asked Dec 16, 2013 at 6:32

6 votes

1 answer

391 views

What is the maximum entropy distribution given values for several quantiles of one sample?

asked Aug 11, 2016 at 1:24

6 votes

1 answer

496 views

Are sample means for quantiles of sorted data unbiased estimators of the true means?

asked Jul 28, 2012 at 1:06

6 votes

2 answers

1k views

Replicate weights and the Current Population Survey

asked Feb 12, 2018 at 5:37

5 votes

1 answer

751 views

Can an estimator of the mean of a distribution with no variance have a variance?

asked Apr 7, 2013 at 23:06

5 votes

2 answers

80 views

Squeezing the juice from a large data set

asked Nov 18, 2016 at 1:53

5 votes

1 answer

1k views

When does the law of large numbers hold for RVs from a distribution with infinite variance?

asked Jun 27, 2016 at 2:33

5 votes

1 answer

41 views

How should I estimate the variance in a sample or population from the sample range?

asked Dec 24, 2022 at 3:46

4 votes

1 answer

54 views

R: How can I represent partially-ordered time series in R?

asked Dec 29, 2019 at 6:23

4 votes

1 answer

60 views

Can regression forecasts of univariate time series be independent (of one another)

asked Dec 20, 2022 at 6:20

4 votes

0 answers

104 views

What do we know about the rate of convergence of the mean of RVs with infinite variance?

asked Jun 27, 2016 at 2:46

4 votes

0 answers

252 views

How to estimate a continuous analog of the (discrete) vector autoregression (VAR) model

asked May 14, 2017 at 3:35

4 votes

2 answers

169 views

Is there a closed-form solution for the tail index of a GB2 distribution?

asked Oct 12, 2017 at 21:47

4 votes

1 answer

2k views

Structural equation models in econometrics vs psychology, political science, etc

asked Jul 18, 2019 at 4:29

3 votes

0 answers

73 views

Measures of central tendency for right-skewed size distributions

asked Mar 7, 2019 at 6:22

3 votes

2 answers

158 views

What is the variance of the mean, conditional on being between two order statistics or quantiles?

asked Sep 1, 2019 at 13:34

3 votes

1 answer

80 views

Good parameter estimates vs good computed moment estimates

asked Feb 21, 2017 at 23:28

3 votes

3 answers

206 views

If the best-fitting distribution has infinite variance, should low observed variance be troubling?

asked Jan 30, 2018 at 5:23

3 votes

1 answer

341 views

Are the Feller-Pareto and the generalized beta distributions really the same?

asked Mar 11, 2018 at 7:57

3 votes

0 answers

38 views

An urn problem: few red balls, many draws (with replacement)

asked Apr 2, 2018 at 2:12

3 votes

1 answer

341 views

How can I fit a polynomial given some prior knowledge but fewer observations than coefficients?

asked Aug 18, 2016 at 1:15

3 votes

2 answers

249 views

What is the expected partial value function really called?

asked Jan 2, 2014 at 23:49

3 votes

0 answers

932 views

What is a good technique for grouping objects based on binary or dichotomous traits?

asked Apr 3, 2014 at 6:55

Stack Exchange Network

107 Questions

Having a conjugate prior: Deep property or mathematical accident?

Does independence imply conditional independence?

What if there is no true data-generating process?

Constructing a continuous distribution to match $m$ moments

Penalized methods for categorical data: combining levels in a factor

Does every log-linear model have a perfectly equivalent logistic regression?

Can I get the parameters of a lognormal distribution from the sample mean & median?

Getting the units right for the Pareto distribution of wealth: x = people, dollars, dollars per person?

What is the maximum entropy distribution given values for several quantiles of one sample?

Are sample means for quantiles of sorted data unbiased estimators of the true means?

Replicate weights and the Current Population Survey

Can an estimator of the mean of a distribution with no variance have a variance?

Squeezing the juice from a large data set

When does the law of large numbers hold for RVs from a distribution with infinite variance?

How should I estimate the variance in a sample or population from the sample range?

R: How can I represent partially-ordered time series in R?

Can regression forecasts of univariate time series be independent (of one another)

What do we know about the rate of convergence of the mean of RVs with infinite variance?

How to estimate a continuous analog of the (discrete) vector autoregression (VAR) model

Is there a closed-form solution for the tail index of a GB2 distribution?

Structural equation models in econometrics vs psychology, political science, etc

Measures of central tendency for right-skewed size distributions

What is the variance of the mean, conditional on being between two order statistics or quantiles?

Good parameter estimates vs good computed moment estimates

If the best-fitting distribution has infinite variance, should low observed variance be troubling?

Are the Feller-Pareto and the generalized beta distributions really the same?

An urn problem: few red balls, many draws (with replacement)

How can I fit a polynomial given some prior knowledge but fewer observations than coefficients?

What is the expected partial value function really called?

What is a good technique for grouping objects based on binary or dichotomous traits?