[R] ks.test
aolinto@bignet.com.br
aolinto at bignet.com.br
Fri Nov 9 15:04:48 CET 2001
Dear R-List members,
I want to check if a set of measurements follows better a
gamma or a lognormal distribution (see data below).
Using shapiro.test I can test for normality (shapiro.test(log
(Lt)).
To test for gamma (and normal) distribution I would use
ks.test but I need to specify its shape and scale. How should
I calculate these values in R?
I tried
> Lt.fit <- glm(Lt ~ 1, family=Gamma)
> gamma.shape(Lt.fit)
Alpha: 18.851759
SE: 3.737472
> ks.test(Lt,"pgamma",18.851759,3.737472)
One-sample Kolmogorov-Smirnov test
data: Lt
D = 0.4063, p-value = 1.358e-07
alternative hypothesis: two.sided
But I'm not sure about it. Using the software Statistica I got
shape = 18.853508 and scale = 3.114434 and
> ks.test(Lt,"pgamma",18.853508,3.114434)
One-sample Kolmogorov-Smirnov test
data: Lt
D = 0.1122, p-value = 0.5553
alternative hypothesis: two.sided
With
> hist(Lt, freq=F)
> curve(dgamma(x,18.851759,3.737472),add=T,col="blue")
> curve(dgamma(x,18.853508,3.114434),add=T,col="red")
I could see that Statistica's parameters fitted better to Lt
distribution. Probably I'm doing something wrong in R.
Thanks in advance,
Antonio Olinto
--------
> sample.dat
Lt
1 60.0
2 79.0
3 48.0
4 57.0
5 84.0
6 57.0
7 57.0
8 43.0
9 58.0
10 63.0
11 66.0
12 53.5
13 60.0
14 62.0
15 41.0
16 65.0
17 50.0
18 85.0
19 68.0
20 46.0
21 45.0
22 59.0
23 53.0
24 37.0
25 61.0
26 106.0
27 47.0
28 49.0
29 56.0
30 42.0
31 54.0
32 87.0
33 51.0
34 78.0
35 45.0
36 49.4
37 67.0
38 84.5
39 54.0
40 57.0
41 49.0
42 57.0
43 59.0
44 83.0
45 43.5
46 42.5
47 59.5
48 65.0
49 44.0
50 49.0
-----------------------------------------------------
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
More information about the R-help
mailing list