# Re: [R] Goodness fit test HELP!

From: Martin Maechler <maechler_at_stat.math.ethz.ch>
Date: Sun 04 Dec 2005 - 06:34:00 EST

>>>>> "Amit" == Amit Kabiri <akabiri@012.net.il>
>>>>> on Sat, 03 Dec 2005 18:43:28 +0200 writes:

```    Amit> If I have a Uniform distribution to check, How can I
Amit> use visual fits? Can I also use in some way the
Amit> qqnorm?

```

yes, in *some* way:

U <- runif(100)

```plot(sort(U))                                     # is already sufficient visually

```

plot(ppoints(U), sort(U)) ; abline(1,1, lty = 3) # is a bit nicer

Amit> Thanks

```    Amit> -----Original Message----- From:
Amit> r-help-bounces@stat.math.ethz.ch
Amit> [mailto:r-help-bounces@stat.math.ethz.ch] On Behalf Of
Amit> Elizabeth Lawson Sent: Friday, November 18, 2005 8:06
Amit> PM To: David Zhao Cc: r-help@stat.math.ethz.ch
Amit> Subject: Re: [R] Goodness fit test HELP!

```

Amit> What about trying a qqplot to see how the distribution     Amit> fits...

Amit> For the normal distribution thta is very stright     Amit> forward, use qqnorm.

Amit> To test gamma distribtution (or any other) do some     Amit> thing like this

Amit> n<-length(data) for(i in 1:n){ prob<-(i-1/3)/(n1/3)     Amit> }

Amit> quantiles<-qgamma(prob,shape=mean(data)/var(data),scale=var(data)/mean(data)     Amit> }

Amit> qqplot(data,quantiles)

Amit> If the distribution is a good for, you should a     Amit> stright line, like wiht a qqnorm plot!

Amit> Good luck!!

Amit> Elizbaeth Lawson

Amit> David Zhao <wzhao6898@gmail.com> wrote: Hi there,

```    Amit> I'm a newbie, plesae bear with me.  I have a dataset
Amit> with about 10000 ~ 30000 data points. Would like fit
Amit> to both Gamma and Normal distribution to see which one
Amit> fits better. How do I do this in R? Or I could do a
Amit> normality test of the data, if it's normal, I then
Amit> will do a normal fit, otherwise, a gamma fit. But
Amit> again, I don't know how to do this either.  Please
Amit> help!

```

Amit> David

Amit> [[alternative HTML version deleted]]

```    Amit> ______________________________________________
Amit> R-help@stat.math.ethz.ch mailing list
```
Amit> http://www.R-project.org/posting-guide.html
```    Amit> ---------------------------------

Amit> 	[[alternative HTML version deleted]]

Amit> ______________________________________________
Amit> R-help@stat.math.ethz.ch mailing list
```
```    Amit> ______________________________________________