User Count Zero

9 votes

2 answers

2k views

Why does adding L1 penalty to R's optim slows things down so much (relative to no penalty or L2)?

asked Aug 15, 2015 at 23:45

8 votes

2 answers

796 views

Do Bayesians use the terminology "statistically significant"?

asked Jun 28, 2016 at 20:22

8 votes

3 answers

3k views

In Stan is there a way to use parameter posterior from old analysis as prior in new analysis?

asked Oct 22, 2016 at 1:05

7 votes

3 answers

3k views

Variance of EM mean estimates in a simple mixture of two normals

asked Feb 19, 2014 at 22:39

5 votes

3 answers

141 views

Is there a term for dividing out a parameter to make model identifiable?

asked Dec 29, 2016 at 23:59

5 votes

1 answer

838 views

Can causal Bayes Nets compute counterfactuals? If so, are they "worse" than structural causal model counterfactuals?

asked Nov 3, 2018 at 0:49

3 votes

1 answer

92 views

When does knowing the causal structure of the data generating process improve supervised learning?

asked Oct 13, 2018 at 14:03

3 votes

0 answers

284 views

Information entropy in a direct graph? [closed]

asked Feb 29, 2016 at 23:57

3 votes

0 answers

384 views

How can I update a Bayesian network model given new data on only a subset of the variables in the original model?

asked Aug 20, 2013 at 12:32

3 votes

1 answer

56 views

Choosing a better model based on the fit of the current model

asked Nov 8, 2015 at 2:05

2 votes

0 answers

739 views

Create a precision matrix and get desired covariance matrix

asked Dec 11, 2015 at 21:32

2 votes

0 answers

377 views

Got an entropy-ish function for a multinomial distribution? Graph theory and Bayes net related

asked Feb 25, 2016 at 23:57

2 votes

1 answer

356 views

How does case-resampling bootstrap work for positive-value estimators?

asked Aug 11, 2016 at 16:00

2 votes

0 answers

219 views

When is logistic regression minimizing under squared error loss the same as maximizing binomial likelihood? [duplicate]

asked May 9, 2015 at 14:21

2 votes

0 answers

560 views

Does R's NLS always fail if model parameters are not identifiable?

asked Jul 20, 2015 at 19:44

2 votes

0 answers

260 views

Understanding my posterior with an uninformative prior with a poisson likelihood. Am I thinking about this correctly?

asked Nov 6, 2016 at 22:22

2 votes

1 answer

461 views

Is there a canonical probability distribution on an ordering/permutation

asked Aug 2, 2018 at 20:05

1 vote

0 answers

1k views

JAGS better than Stan for fitting CPDs in a Bayesian Network?

asked Nov 22, 2016 at 19:17

1 vote

1 answer

38 views

Choosing one parameterization over another

asked Dec 27, 2016 at 19:53

1 vote

0 answers

340 views

Looking for derivation of Shannon entropy of binomial distribution

asked Apr 15, 2016 at 23:47

1 vote

1 answer

89 views

$X \sim$ Poisson$(λ)$. What is the distribution of $X/c$? $(c > 0)$

asked Nov 5, 2016 at 0:32

1 vote

1 answer

54 views

Need GLM ideas for nonlinear biochemical model

asked Aug 2, 2015 at 0:59

1 vote

0 answers

88 views

Tried to overfit a Bayes net, but mean prediction error is worse than learned network?

asked Nov 1, 2013 at 22:48

1 vote

0 answers

60 views

Need a statistic for comparing "strength" of Markov blankets in a Bayesian network

asked Mar 12, 2015 at 22:15

1 vote

1 answer

530 views

Neural networks: how can convex optimization produce different weights each time?

asked Apr 11, 2015 at 8:13

1 vote

1 answer

76 views

How can I make this biological relation into a glm model?

asked May 7, 2015 at 22:00

1 vote

0 answers

33 views

How do I model of Y| X, M when data has only X and Y not M, an upper bound on Y?

asked Sep 27, 2016 at 23:53

0 votes

0 answers

166 views

How to construct prior for two variables based on known distribution of their product?

asked Oct 11, 2016 at 22:44

0 votes

0 answers

192 views

Are there posterior predictive check "point-values" in Bayesian modeling?

asked Jul 25, 2017 at 15:23

0 votes

0 answers

32 views

Logit analysis of binomial response in contigency tables with > 2 categorical predictors

asked Aug 23, 2019 at 14:12

Stack Exchange Network

30 Questions

Why does adding L1 penalty to R's optim slows things down so much (relative to no penalty or L2)?

Do Bayesians use the terminology "statistically significant"?

In Stan is there a way to use parameter posterior from old analysis as prior in new analysis?

Variance of EM mean estimates in a simple mixture of two normals

Is there a term for dividing out a parameter to make model identifiable?

Can causal Bayes Nets compute counterfactuals? If so, are they "worse" than structural causal model counterfactuals?

When does knowing the causal structure of the data generating process improve supervised learning?

Information entropy in a direct graph? [closed]

How can I update a Bayesian network model given new data on only a subset of the variables in the original model?

Choosing a better model based on the fit of the current model

Create a precision matrix and get desired covariance matrix

Got an entropy-ish function for a multinomial distribution? Graph theory and Bayes net related

How does case-resampling bootstrap work for positive-value estimators?

When is logistic regression minimizing under squared error loss the same as maximizing binomial likelihood? [duplicate]

Does R's NLS always fail if model parameters are not identifiable?

Understanding my posterior with an uninformative prior with a poisson likelihood. Am I thinking about this correctly?

Is there a canonical probability distribution on an ordering/permutation

JAGS better than Stan for fitting CPDs in a Bayesian Network?

Choosing one parameterization over another

Looking for derivation of Shannon entropy of binomial distribution

$X \sim$ Poisson$(λ)$. What is the distribution of $X/c$? $(c > 0)$

Need GLM ideas for nonlinear biochemical model

Tried to overfit a Bayes net, but mean prediction error is worse than learned network?

Need a statistic for comparing "strength" of Markov blankets in a Bayesian network

Neural networks: how can convex optimization produce different weights each time?

How can I make this biological relation into a glm model?

How do I model of Y| X, M when data has only X and Y not M, an upper bound on Y?

How to construct prior for two variables based on known distribution of their product?

Are there posterior predictive check "point-values" in Bayesian modeling?

Logit analysis of binomial response in contigency tables with > 2 categorical predictors