Re: [R] Minimization problem
On Thu, 2010-05-20 at 01:35 -0700, Fred wrote: Dear R users, I am trying to minimize two function simultaneously in R, function(x) minimize x[1],x[2],x[3] mean(distribution(x1,x2,x3) ) - observed mean std(distribution(x1,x2,x3)) - observed std What I want to achieve is that simulated mean and standard deviation of distribution related to x1 x2 x3 would be close to observed mean and observed standard deviation. is there any function in R can reach this? Thank you for the help first . F. Hi! Do you need use optim, something like this test - function(parameters){ m.error - mean(distribution(x1,x2,x3) ) - observed mean m.sd - std(distribution(x1,x2,x3)) - observed std res - cbind(m.error,sd.error) return(res) } -- Bernardo Rangel Tura, M.D,MPH,Ph.D National Institute of Cardiology Brazil __ 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.
Re: [R] Minimization problem
That would not work in optim. The objective function must take in a vector of parameters and return a scalar value. Your objective function does not return a scalar. This would work: test - function(x, ...){ m.error - mean(distribution(x)) - observed.mean sd.error - std(distribution(x)) - observed.std res - c(m.error, sd.error) return(sum(res^2)) # returns a scalar } Ravi. -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of Bernardo Rangel Tura Sent: Thursday, May 20, 2010 6:02 AM To: r-help@r-project.org Subject: Re: [R] Minimization problem On Thu, 2010-05-20 at 01:35 -0700, Fred wrote: Dear R users, I am trying to minimize two function simultaneously in R, function(x) minimize x[1],x[2],x[3] mean(distribution(x1,x2,x3) ) - observed mean std(distribution(x1,x2,x3)) - observed std What I want to achieve is that simulated mean and standard deviation of distribution related to x1 x2 x3 would be close to observed mean and observed standard deviation. is there any function in R can reach this? Thank you for the help first . F. Hi! Do you need use optim, something like this test - function(parameters){ m.error - mean(distribution(x1,x2,x3) ) - observed mean m.sd - std(distribution(x1,x2,x3)) - observed std res - cbind(m.error,sd.error) return(res) } -- Bernardo Rangel Tura, M.D,MPH,Ph.D National Institute of Cardiology Brazil __ 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. __ 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.
Re: [R] Minimization Problem
I'm not sure that it comes across as a matrix problem in the sense of involving matrix algebra, but it's also possible that I don't understand what you are saying. Seems that an appropriate application of: ?abs ?max ?sapply ?which.min ... ought to do the trick. If you were hoping for the name of the column vector, then may also need to look at: ?names -- David Winsemius On Jan 18, 2009, at 7:17 AM, glenn wrote: Dear All, Could someone give me some pointers (just a guide as to what functions I need to look at would be fine) as to how I go about this simple problem please; The problem looks like this; Choose x1 to x4 such that you minimize the MAXIMUM ABSOLUTE value returned in the vector result of this matrix problem; [x1,x2,x3,x4].[ -0.38 -0.52 0.68-0.29 ] [ -0.39 -0.42-0.12 0.49 ] [ -0.40 -0.20-0.45 0.26 ] [ -0.390.03-0.39-0.36 ] I hope the formatting comes out ok , I did this on a Mac ! Glenn [[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. __ 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.
Re: [R] Minimization Problem
Hi Glenn, perhaps I'm misunderstanding your problem, but it seems to me that the zero vector (yielding a zero vector after the multiplication) will happily minimize everything you want. Apart from that, you may want to look at the Optimization Task View on CRAN. Good luck, Stephan glenn schrieb: Dear All, Could someone give me some pointers (just a guide as to what functions I need to look at would be fine) as to how I go about this simple problem please; The problem looks like this; Choose x1 to x4 such that you minimize the MAXIMUM ABSOLUTE value returned in the vector result of this matrix problem; [x1,x2,x3,x4].[ -0.38 -0.52 0.68-0.29 ] [ -0.39 -0.42-0.12 0.49 ] [ -0.40 -0.20-0.45 0.26 ] [ -0.390.03-0.39-0.36 ] I hope the formatting comes out ok , I did this on a Mac ! Glenn [[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. __ 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.