[R] threshold distribution

Bernardo Rangel Tura tura at centroin.com.br
Sun Apr 5 11:40:06 CEST 2009


On Sun, 2009-04-05 at 01:13 -0400, jim holtman wrote:
> Here is what I get from using 'fitdistr' in R to fit to a lognormal.
> The resulting density plot from the distribution seems to be a reason
> match to the data.
> 
> > x <- scan()
> 1: 0.80010 0.72299 0.69893 0.99597 0.89200 0.69312 0.73613 1.13559
> 9: 0.85009 0.85804 0.73324 1.04826 0.84002
> 14: 1.76330 0.71980 0.89416 0.89450 0.98670 0.83571 0.73833 0.66549
> 22: 0.93641 0.80418 0.95285 0.76876 0.82588
> 27: 1.09394 1.00195 1.14976 0.80008 1.11947 1.09484 0.81494 0.68696
> 35: 0.82364 0.84390 0.71402 0.80293 1.02873
> 40:
> Read 39 items
> > plot(density(x))
> > library(MASS)
> > fitdistr(x, 'lognormal')
>      meanlog        sdlog
>   -0.13480636    0.19118861
>  ( 0.03061468) ( 0.02164785)
> > lines(dlnorm(x, -.1348, .1911), col='red')

Hi Jim

I agree with your solution but my plot result not fine.
I obtain same result.
> fitdistr(x, 'lognormal')
     meanlog        sdlog   
  -0.13480636    0.19118861 
 ( 0.03061468) ( 0.02164785)

In plot when I use points (blue) and curve (green) the fit o lognormal
and density(data) is fine but when I use lines (red) the fit is bad (in
attach I send a PDF of my output)

Do you know why this  happen?

Thanks in advance

-- 
Bernardo Rangel Tura, M.D,MPH,Ph.D
National Institute of Cardiology
Brazil

P.S. my script is

x <- scan()
0.80010 0.72299 0.69893 0.99597 0.89200 0.69312 0.73613 1.13559
0.85009 0.85804 0.73324 1.04826 0.84002
1.76330 0.71980 0.89416 0.89450 0.98670 0.83571 0.73833 0.66549
0.93641 0.80418 0.95285 0.76876 0.82588
1.09394 1.00195 1.14976 0.80008 1.11947 1.09484 0.81494 0.68696
0.82364 0.84390 0.71402 0.80293 1.02873

require(MASS)
fitdistr(x, 'lognormal')
pdf("adj.pdf")
plot(density(x))
lines(dlnorm(x, -.1348, .1911), col='red')
points(x,dlnorm(x, -.1348, .1911), col='blue')
curve(dlnorm(x, -.1348, .1911), col='green',add=T)
dev.off()


-------------- next part --------------
A non-text attachment was scrubbed...
Name: adj.pdf
Type: application/pdf
Size: 15127 bytes
Desc: not available
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20090405/a19ba224/attachment-0002.pdf>


More information about the R-help mailing list