Good morning, in my work I need to study data distributions and so I need to fit the experimental distribution by theoretical curves such as normal, Poison, binomial and so on. I'd like to know, given a vector of data, for example x<-rnorm (1000, 10) if they follow a normal distribution. I'd like to do a fit (to estimate the parameters of the theoretical distribution) and then a goodness test. Can you suggest me any R package or manuals about this issue? The documentation on the R-guide isn't sufficient to me. Thank you of your help, Angela
On Wed, 22 Dec 2004 15:12:48 +0100 Angela Re wrote:> Good morning, > in my work I need to study data distributions and so I need to fit > the experimental distribution by theoretical curves such as normal, > Poison, binomial and so on. I'd like to know, given a vector of data, > for example > > x<-rnorm (1000, 10) > > if they follow a normal distribution. I'd like to do a fit (to > estimate the parameters of the theoretical distribution) and then a > goodness test.For fitting distributions (via ML) look at ?fitdistr in package MASS. Goodness-of-fit testing after the estimation of parameters might not be straightforward, but there are several solutions for testing normality in package nortest. Also look at ?shapiro.test and ?goodfit in package vcd. All purpose gof tests are also available, see ?ks.test and ?chisq.test. Z> Can you suggest me any R package or manuals about this issue? The > documentation on the R-guide isn't sufficient to me. > Thank you of your help, Angela > > ______________________________________________ > R-help at stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! > http://www.R-project.org/posting-guide.html >
See http://www.statsoft.com/textbook/stdisfit.html There are several approaches you can use - Chi-square, Q-Q plots, P-P plots, various tests (Kolmogorov-Smirnov, Shapiro-Wilks' W) etc. HTH, b. -----Original Message----- From: Angela Re Sent: Wednesday, December 22, 2004 9:13 AM To: r-help at stat.math.ethz.ch Subject: [R] how to fit in R Good morning, in my work I need to study data distributions and so I need to fit the experimental distribution by theoretical curves such as normal, Poison, binomial and so on. I'd like to know, given a vector of data, for example x<-rnorm (1000, 10) if they follow a normal distribution. I'd like to do a fit (to estimate the parameters of the theoretical distribution) and then a goodness test. Can you suggest me any R package or manuals about this issue? The documentation on the R-guide isn't sufficient to me. Thank you of your help, Angela ______________________________________________ R-help at stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html