sigalit mangut-leiba
2007-Oct-17 08:03 UTC
[R] how to repeat the results of a generated probabilities
hello, I want to simulate 200 times the mean of a joint probability (y1) and 200 times the mean of another joint distribution (y2), that is I'm expecting to get 200 means of y1 and 200 means of y2. y1 and y2 are probabilities that I calculate from the marginal prob. (z1 and z2 respectively) multiple by the conditional prob. (x1 and x2 respectively), which I generaterd from the binomial distribution. here are the results for one mean of y1 and one mean of y2: > y1 <- 0> for (i in 1:200){+ z1[i] <- rbinom(1, 1, .6) + x1[i] <- rbinom(1, 1, .95) + y1[i] <- z1[i]*x1[i] + }> mean(y1)[1] 0.605> y2 <- 0 > for (j in 1:200){+ z2[j] <- rbinom(1, 1, .6) + x2[j] <- rbinom(1, 1, .8) + y2[j] <- z2[j]*x2[j] + }> mean(y2)[1] 0.515>and this I want to simulate 200 time, in order to see the distribution of y1 and the distribution of y2. How can I repeat it and see all the results? (all the 200 means of y1 and y2) Thank you, Sigalit. [[alternative HTML version deleted]]
sigalit mangut-leiba
2007-Oct-17 19:12 UTC
[R] how to repeat the results of a generated probabilities
It does! thank you very much. Sigalit. On 10/17/07, Peter McMahan <peter.mcmahan@gmail.com> wrote:> > There are two things that will help you out. > > First, instead of getting the individual z1 and x1 one at a time and > then multiplying them you can call rbinom(200,1,p) to get a vector of > 200 independent draws. So you could replace your for loop with: > > z1 <- rbinom(200,1,.6) > x1 <- rbinom(200,1,.95) > y1 <- z1*x1 > > R will multiply vectors element-wise unless you use matrix multiply > so this will get you the same result much faster and easier. > > Second, if you don't care about saving the z1 and x1 vectors, the > replicate() function ("help(replicate)") can run the "experiment" any > number of times and return it to you as yet another vector. These two > lines > > yMeans1 <- replicate(200,mean(rbinom(200,1,.6)*rbinom(200,1,.95))) > yMeans2 <- replicate(200,mean(rbinom(200,1,.6)*rbinom(200,1,.8))) > > will give you the end results you're looking for, with yMeans1 being > a vector of means of 200 independent runs of the full 200 draws. The > first argument is the number of times to replicate, the second is the > expression you want to run multiple times. > > Hope this helps. > > On Oct 17, 2007, at 3:03 AM, sigalit mangut-leiba wrote: > > > hello, > > I want to simulate 200 times the mean of a joint probability (y1) > > and 200 > > times the mean of another joint distribution (y2), > > that is I'm expecting to get 200 means of y1 and 200 means of y2. > > y1 and y2 are probabilities that I calculate from the marginal > > prob. (z1 and > > z2 respectively) multiple by the conditional prob. (x1 and x2 > > respectively), > > which I generaterd from the binomial distribution. > > here are the results for one mean of y1 and one mean of y2: > > > >> y1 <- 0 > >> for (i in 1:200){ > > + z1[i] <- rbinom(1, 1, .6) > > + x1[i] <- rbinom(1, 1, .95) > > + y1[i] <- z1[i]*x1[i] > > + } > >> mean(y1) > > [1] 0.605 > >> y2 <- 0 > >> for (j in 1:200){ > > + z2[j] <- rbinom(1, 1, .6) > > + x2[j] <- rbinom(1, 1, .8) > > + y2[j] <- z2[j]*x2[j] > > + } > >> mean(y2) > > [1] 0.515 > >> > > and this I want to simulate 200 time, in order to see the > > distribution of y1 > > and the distribution of y2. > > How can I repeat it and see all the results? (all the 200 means of > > y1 and > > y2) > > Thank you, > > Sigalit. > > > > [[alternative HTML version deleted]] > > > > ______________________________________________ > > R-help@r-project.org mailing list > > https://stat.ethz.ch/mailman/listinfo/r-help > > PLEASE do read the posting guide http://www.R-project.org/posting- > > guide.html > > and provide commented, minimal, self-contained, reproducible code. > >[[alternative HTML version deleted]]
Hi all, Somebody knows how to compare 2 correlation matrices using R? Any suggestion would be appreciate. Thanks in advance, Jorge
I looked at this a while ago - what are you trying to do? ie what are you correlating? ________________________________ From: r-help-bounces at r-project.org on behalf of Jorge Ivan Velez Valbuena Sent: Thu 18/10/2007 12:20 AM To: r-help Subject: [R] Correlation matrices comparison Hi all, Somebody knows how to compare 2 correlation matrices using R? Any suggestion would be appreciate. Thanks in advance, Jorge ______________________________________________ R-help at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.