otsvetkova
2012-Jul-22 07:31 UTC
[R] Adding gamma and 3-parameter log normal distributions to L-moments ratio diagram lmrd()
How to adapt this piece of code but for: - gamma distribution - 3 parameter log normal More specifically, where can I find the specification of the parameter (lmom) for pelgam() and pelln3()? Lmom package info just gives: pelgam(lmom), lelln3(lmom), where lmom is a numeric vector containing the L-moments of the distribution or of a data sample. # Draw an L-moment ratio diagram and add a line for the # Weibull distribution lmrd() weimom <- sapply( seq(0, 0.9, by=0.01), function(tau3) lmrwei(pelwei(c(0,1,tau3)), nmom=4) ) lmrdlines(t(weimom), col='darkgreen', lwd=2) source: http://rgm2.lab.nig.ac.jp/RGM2/func.php?rd_id=lmom:lmrdpoints Thank you. -- View this message in context: http://r.789695.n4.nabble.com/Adding-gamma-and-3-parameter-log-normal-distributions-to-L-moments-ratio-diagram-lmrd-tp4637345.html Sent from the R help mailing list archive at Nabble.com.
J. R. M. Hosking
2012-Jul-23 14:02 UTC
[R] Adding gamma and 3-parameter log normal distributions to L-moments ratio diagram lmrd()
On 2012-07-22 03:31, otsvetkova wrote:> How to adapt this piece of code but for: - gamma distribution - 3 parameter > log normal > More specifically, where can I find the specification of the parameter > (lmom) for pelgam() and pelln3()? > > Lmom package info just gives: pelgam(lmom), lelln3(lmom), where lmom is a > numeric vector containing the L-moments of the distribution or of a data > sample. > > # Draw an L-moment ratio diagram and add a line for the > # Weibull distribution > lmrd() > weimom<- sapply( seq(0, 0.9, by=0.01), > function(tau3) lmrwei(pelwei(c(0,1,tau3)), nmom=4) ) > lmrdlines(t(weimom), col='darkgreen', lwd=2) > > source: http://rgm2.lab.nig.ac.jp/RGM2/func.php?rd_id=lmom:lmrdpoints > > Thank you.lmrd() by default draws lines for the Pearson type III distribution (which is a 3-parameter form of the gamma distribution) and the generalized normal distribution (a differently parametrized version of the 3-parameter lognormal distribution). Is this what you want? If not, I don't know what else you want to know about "the specification of the parameter (lmom)". J. R. M. Hosking