Skip to content

Shapiro-Wilk normality test

5 messages · Bert Gunter, Hard Core, Kevin Wright

#
1. Is this homework?
2. This is not an R question. Post to a statistical list list
stats.stackexchange.com for such questions.
3. If this is not homework, I suggest you consult a local statistical
expert, as your statistics knowledge appears minimal.
4. No variable is **ever** normally distributed.
5. Tests of normality are a waste of time (this is **my** opinion only).

Cheers,
Bert
On Fri, Nov 2, 2012 at 7:05 AM, Hard Core <gioxc at hotmail.it> wrote:

  
    
#
my knowledge of statistical is not minimal .. 
i was just a little bit confused because on wikipedia i read that the value
of SW test 0 < w < 1. When w is very little then you're supposed to refuse
null hypotesis. So in this case, considering the value of w i should accept
null hypotesis, but p-value tells me the contrary.  So i thought that i
missed something in the formula.

That's not homework ..





--
View this message in context: http://r.789695.n4.nabble.com/Shapiro-Wilk-normality-test-tp4648236p4648276.html
Sent from the R help mailing list archive at Nabble.com.
#
Dear Kewin, this is just my issue. Take a look at this

<http://r.789695.n4.nabble.com/file/n4648278/Rplot02.png> 

I have skewness = 0.16 and kurtosis = 3.3 ... This one seem to me normally
distribued but both Shapiro-Wilk and Jarque-Bera test make me refuse null
hypotesis of normality



--
View this message in context: http://r.789695.n4.nabble.com/Shapiro-Wilk-normality-test-tp4648236p4648278.html
Sent from the R help mailing list archive at Nabble.com.