Comments for John Myles White

Comment on Using Complex Numbers in R by Micke

Micke — Tue, 24 Jan 2012 09:01:49 +0000

I would like to point out that the statement

Mod(z) == z * Conj(z)

is only true when Mod(z)==1. In general it should be

Mod(z)^2 == z * Conj(z)

Comment on MCMC Diagnostics in R with the coda Package by John Myles White

John Myles White — Mon, 23 Jan 2012 00:13:59 +0000

I’m glad they were of so much use.

Comment on MCMC Diagnostics in R with the coda Package by dan doak

dan doak — Fri, 20 Jan 2012 23:20:20 +0000

thanks very much for your two posts on rjags. they are Far more helpful than anything else i could on getting started with rjags — i really appreciate your time in putting this info up on the web
.
dan

Comment on Pearson vs. Spearman Correlation Coefficients by Suz

Suz — Tue, 17 Jan 2012 04:22:09 +0000

Jen,
Did you have any luck? I’m trying to find this too.
Suz

Comment on Pearson vs. Spearman Correlation Coefficients by Ajay

Ajay — Wed, 21 Dec 2011 02:43:39 +0000

Hi JOhn, thanks for explanation above. Hope you can help me with below:-

What would be the best to identify correlation between sales of brands available in supermarkets?
Note:- I don’t have transaction data, just the total sales in a given week for two years. I am planning to time series analysis to find out correlation between brand hitting peaks at the same time and brands getting affected by it. Planning to Pearson first and then Spearman or Kindall later once the ranking is defined. Any thoughts on this would be much appreciated.

Comment on ProjectTemplate by yolio

yolio — Sat, 10 Dec 2011 21:34:38 +0000

I have a little nitpicky comment: I would call the “graphs” directory “figures” instead. A lot of images that get produced aren’t really graphs, per se.

Comment on Using JAGS in R with the rjags Package by John Myles White

John Myles White — Thu, 08 Dec 2011 20:18:47 +0000

@Filipe, I unfortunately don’t have any experience getting JAGS working on a Windows machine. Perhaps move over to WinBUGS?

@Jake, tau (really sigma) is controlling the variance of the linear model. I think of this as a quantity you very much want to use a wide prior for so that the data is in full control, but I don’t have a strong intuition about what will happen if you set it to a lower value through the prior. Try giving sigma a uniform(0, 0.01) prior and see what happens. I’d be interested to know.

Comment on Using JAGS in R with the rjags Package by Jake

Jake — Thu, 08 Dec 2011 20:12:12 +0000

Hey John,

Nice post. Quick question. I am just now getting into some bayesian modeling of simple linear regressions but I have data that i want to use to inform my model. My question is this: I understand how to inform the alpha and beta, that makes clear sense to me, but when do you/how do you know to place informed priors on tau? I guess I don’t really understand what tau is doing.

Any comments would be great.

thanks,

Jake

Comment on Using JAGS in R with the rjags Package by Filipe Ferminiano

Filipe Ferminiano — Thu, 08 Dec 2011 10:24:39 +0000

Nice tutorial, but I’m having trouble with rjags in Windows 7. R crushes when I try to run the line:
jags <- jags.model('example1.bug', data = list('x' = x,'N' = N),n.chains = 4,n.adapt = 100)
I'm using a 64 bit Windows 7 version. I have already updated to the last R version, but it failed in the same way.
I tried run this tutorial in a Windows XP machine and it worked well.
Can you help me?

Comment on Implementing Push and Pop in R by Iain S

Iain S — Sat, 03 Dec 2011 17:42:10 +0000

Kevin’s solution works nicely, except if you try pushing a string onto a vector. In that case, you need to enclose the item in quotes, as below:

push = function(vec, item)
{
vec = substitute(vec)

if (is(item, “character”)) {
eval.parent(parse(text = paste(vec, ‘ <- c(', vec, ', "', item, '")', sep = '')), n=1)
} else {
eval.parent(parse(text = paste(vec, ' <- c(', vec, ', ', item, ')', sep = '')), n=1)
}
}

Comment on ProjectTemplate News by John Myles White

John Myles White — Wed, 23 Nov 2011 22:59:10 +0000

Glad to hear it, RK.

Comment on Text Processing in R by Priya

Priya — Wed, 23 Nov 2011 10:06:02 +0000

Nice post very helpful…
Hi eric u can do that by searching for the digit in ur string first.. \d+..
and then find out its indice and u may use substr and split the string into 2 parts…

Comment on ProjectTemplate News by RK

RK — Tue, 22 Nov 2011 02:12:57 +0000

Hi,
Your library is tremendously useful in organizing R projects. I have just started using your package and it has helped me a LOT in many ways. Thanks a ton for writing the package.

Comment on Emotional Determinism by aware

aware — Tue, 08 Nov 2011 06:15:04 +0000

I know that feel bro.

Comment on Pearson vs. Spearman Correlation Coefficients by Matias

Matias — Thu, 03 Nov 2011 21:29:44 +0000

Very good forum, so … my question now is. It is correct to say “there is a correletion between the variable X and variable Y” —–> if i had a significative Rho of a spearman ?¿. Or which would be the best way to present my results in a conference.

Thank you very much !