search

certifiedwaif / phd

5 stars 0 forks source link

Lower bound for multivariate ZIP model does not monotonically increase. #6

Closed certifiedwaif closed 10 years ago

certifiedwaif commented 10 years ago

It usually goes up, but not always. I'll post more details when I get home.

certifiedwaif commented 10 years ago

After re-calculating the lower bound expressions, I found some errors, and some omissions. For the two test cases I tried, the lower bound now monotonically increases when there's a random intercept term. It doesn't monotonically increase when there are only fixed effects.

But still, progress!

JohnOrmerod commented 10 years ago

Dear Mark, Great! Discuss sometime tomorrow? Otherwise Friday is good. Kind regards, John

From: Mark Greenaway [notifications@github.com] Sent: Monday, August 25, 2014 5:39 PM To: certifiedwaif/phd Subject: Re: [phd] Lower bound for multivariate ZIP model does not monotonically increase. (#6)

After re-calculating the lower bound expressions, I found some errors, and some omissions. For the two test cases I tried, the lower bound now monotonically increases when there's a random intercept term. It doesn't monotonically increase when there are only fixed effects.

But still, progress!

— Reply to this email directly or view it on GitHubhttps://github.com/certifiedwaif/phd/issues/6#issuecomment-53235884.

certifiedwaif commented 10 years ago

Sure!

On Mon, Aug 25, 2014 at 7:37 PM, JohnOrmerod notifications@github.com wrote:

Dear Mark, Great! Discuss sometime tomorrow? Otherwise Friday is good. Kind regards, John

From: Mark Greenaway [notifications@github.com] Sent: Monday, August 25, 2014 5:39 PM To: certifiedwaif/phd Subject: Re: [phd] Lower bound for multivariate ZIP model does not monotonically increase. (#6)

After re-calculating the lower bound expressions, I found some errors, and some omissions. For the two test cases I tried, the lower bound now monotonically increases when there's a random intercept term. It doesn't monotonically increase when there are only fixed effects.

But still, progress!

— Reply to this email directly or view it on GitHub< https://github.com/certifiedwaif/phd/issues/6#issuecomment-53235884>.

— Reply to this email directly or view it on GitHub https://github.com/certifiedwaif/phd/issues/6#issuecomment-53245187.