Comments for James Keirstead

Comment on Positive and negative politeness by mrfive

mrfive — Mon, 26 Dec 2011 14:09:22 +0000

this is very usefull…
I need more explanation about politeness theory

Comment on Free Meta font by fatima

fatima — Sat, 12 Nov 2011 09:43:19 +0000

nice fonts

Comment on Changing the fonts in R plots by Drew

Drew — Wed, 26 Oct 2011 11:53:21 +0000

For people like me, who found this useful and then wanted to know about setting fonts in pdf files: there’s a nice post about that here: http://stackoverflow.com/questions/1395323/fonts-in-r-plots#_=_

Comment on How to: Binomial regression models in R by Natalia

Natalia — Mon, 19 Sep 2011 18:15:46 +0000

Hello, I have a doubt about binomial models: I did a binomial glm based on an experiment, and my response variable es female bird agression; females in my experimental group were agressive and females in the control group were not. when I run the model, theres some mistake, because all the 0′s are on the control group and all the 1′s are in the experimental group….I think it is a quasicomplete separation problem…..¿what can I do to run a binomial glm with data like this??

thanx a lot

Comment on How to: Binomial regression models in R by James Keirstead

James Keirstead — Sun, 18 Sep 2011 17:40:55 +0000

Not off the top of my head, no. But if any good book on microsimulation should describe this general approach.

Comment on How to: Binomial regression models in R by John

John — Sat, 17 Sep 2011 23:02:49 +0000

Hi James,

Thanks for the post. Do you have any references for this approach to draw a random number in order to determine the binary response?

Comment on Positive coefficient regression in R by James Keirstead

James Keirstead — Tue, 09 Aug 2011 05:59:09 +0000

Hi Eric

Yes, sorry I should have clarified. All of the activities represent things like “work”, “education” or “shopping” which you would expect to have positive impact. Also the figure shows a commercial zone, i.e. where the activities are taking place, and not a domestic zone where demand might go down because people have left to do something else.

James

Comment on Positive coefficient regression in R by Erolgo

Erolgo — Tue, 09 Aug 2011 04:39:54 +0000

Er, … it’s OK isn’t it, to have negative coefficients on some activities? Time spent in that activity means less time spent in another, more costly, activity, hence represents a saving.
Not looking at the specific activities in your list, you’d imagine that activities such as “turning off lights” or “picnicing by a river” could conceivably have a lower than average electricity demand, and thus have a negative coefficient.
Or have I missed the point again?
Eric

Comment on Slopegraphs in R by More on Presidential Rhetoric « YourMorals.Org Moral Psychology Blog

More on Presidential Rhetoric « YourMorals.Org Moral Psychology Blog — Tue, 26 Jul 2011 22:51:30 +0000

[...] credit James Kiersted whose code I liberally borrowed from in constructing the figures above. See this post at his blog for more. Posted in differences between republicans and democrats, moral [...]

Comment on Slopegraphs in R by Bob Muenchen

Bob Muenchen — Mon, 18 Jul 2011 13:05:53 +0000

Nice work! I think there’s still a bit of tweaking to do. In Tufte’s version, the final Thyroid value 95 appears even or slightly higher than the first 96 value due to (I think) an optical illusion caused by the lines dipping to the 3rd value, 94. If you lay a ruler across the string, “Thyroid 96 96″ the final 95 is slightly below it. But on your version, that approach shows the 95 is actually above the 96. Even the first two 96′s are not level. But you’re really close & I love the plot! It has been years since I read that book & I had forgotten all about it.

Cheers,
Bob