[R] how to interpret KS test

[Cedric Laczny]

Hi Izidine,

On Friday, 20. August 2010 15:19:24 Izidine Pinto wrote:
> Dear R users
> I am using KS test to compare two different distribution for the same
> variable (temperature)  for two different time periods.
> H0: the two distributions are equal
> H1: the two distributions are different
> 
> ks.test (temp12, temp22)
> 
>     Two-sample Kolmogorov-Smirnov test
> 
> data:  temp12 and temp22

This tells you where the data comes from that was used in this test, basically 
the program-variable you assigned the values to.

> D = 0.2047, p-value < 2.2e-16

D represents the value of the test-statistic (difference), so the KS-statistic 
and the p-value represents the likelihood of observing this particular value 
of D, or a "more extreme" value by pure chance.

> alternative hypothesis: two-sided
> 
> Warning message:
> In ks.test(temp12, temp22) : cannot compute correct p-values with ties

This tells you, that ties occured when performing the test. A "tie" means that 
two or more samples had the same value. The problem with ties is basically, 
that it results in difficulties calculating the variance of your variable. Some 
tests account for this by calculating exact p-values based on permutations, 
which is, especially for larger sample sizes, computationally expensive.

> I don't rally know how to interpret the  output from R.

I don't want to judge your skills on statistics. However, it seems to me, that 
you are not completely familiar with the concept of hypothesis testing, the 
applied statistical tests and the interpretation of the results.
Understanding what the test statistic and p-value (confidence interval etc.) 
are is fundamental for any subsequent steps, especially the interpretation of 
the calculated results. I had to learn this personally only recently ;)
After all, R is just a tool, but the concepts behind the methods it offers must 
be understood separately as they are basic statistics most of the time.

> Any help is appreciated.
> 
> **

Hope that makes it a bite more clear. Otherwise, please feel free to ask 
further :)

> 
> 	[[alternative HTML version deleted]]
> 

Best,

Cedric

> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html and provide commented,
> minimal, self-contained, reproducible code.