
Re: Skewness and kurtosis pvalues
Posted:
May 25, 2013 12:36 PM


"Cristiano" wrote in message news:knq8n3$gmc$1@dontemail.me...
On 25/05/2013 12:43, Cristiano wrote: > If someone can confirm that the following procedure is good, I can stop > asking and I can start the simulation:
Better idea: I checked my simulation with a countersimulation. I count how many calculated skewness fall beyond the critical values calculated with my simulation. The results are very good.
I really don't know how that site gets those critical values.
Cristiano
=======================================================
Probably, that site knows what a "twosided test" means whereas, judging by your description of simulation for the skewness, you do not. The simplest change to your procedure would be to use the absolute value of the calculated skewness, since that is the test statistic for a twosided test in this case. On the webpage, "alpha" is the total area of the two tails, not just one tail.
You also said "Step 2a: for the kurtosis I need 2 critical values, but for the skewness do I really need 2 critical values?". You do need two critical values for the raw skewness, but for symmetric distributions you know that these are related in a simple way. If you were working out a test of skewness for some nonsymmetric distribution, as is certainly possible, there would be nonsymmetric lower and upper limits for a twosided test.
David Jones

