[R] fitting distribution

eliza botto eliza_botto at hotmail.com
Tue Sep 30 00:38:21 CEST 2014


Dear UseRs,
I have a dataset in the following form.
> dput(Da)
structure(c(0.0238095238095238, 0.0476190476190476, 0.0714285714285714, 0.0952380952380952, 0.119047619047619, 0.142857142857143, 0.166666666666667, 0.19047619047619, 0.214285714285714, 0.238095238095238, 0.261904761904762, 0.285714285714286, 0.30952380952381, 0.333333333333333, 0.357142857142857, 0.380952380952381, 0.404761904761905, 0.428571428571429, 0.452380952380952, 0.476190476190476, 0.5, 0.523809523809524, 0.547619047619048, 0.571428571428571, 0.595238095238095, 0.619047619047619, 0.642857142857143, 0.666666666666667, 0.69047619047619, 0.714285714285714, 0.738095238095238, 0.761904761904762, 0.785714285714286, 0.80952380952381, 0.833333333333333, 0.857142857142857, 0.880952380952381, 0.904761904761905, 0.928571428571429, 0.952380952380952, 0.976190476190476, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190, 200, 210, 220, 230, 240, 250, 260, 270, 280, 290, 300, 310, 320, 330, 340, 350, 360, 370, 380, 390, 400, 410, 420, 430, 440, 450, 460, 470, 480, 490, 500), .Dim = c(41L, 2L), .Dimnames = list(NULL, c("P", "RP")))

Column number 1 is the probability of non execeedance and column 2 is return period of the event. I plotted the dataset in the following way
>plot(Da[,2],Da[,1],log="x")
I now want to fit gumbel distribution, LN3 and GEV distributions on the data and plot them on the same window to see which distribution actually fits good to the data. I know there are other ways of determining best fitted distribution in R but I just wanted to know that if I can actually plot it on actual distribution or not. I want it to try for my thesis.
Thankyou very much in advance
Eliza
 		 	   		  
	[[alternative HTML version deleted]]



More information about the R-help mailing list