Normality test and intrpretation of swilk shapiro test

[Fadi Ansar]

Dear All
Can help me confusions i had while checking the noramlity of data.I have checked with histogram and qq polt and both confirmed that my data is not normal.This doubt is very basic and some one might feel what kind of qeustion is this.I am very new to this area and I couldnt find an answer for this I have also run shapiro wilk test and giving the results below

 
Variable
Obs
W
V
z
Prob>z


    
Z score
1,169
    0.675
  236.638
   13.623
    0.000


    
NPL
971
    0.133
  532.649
   15.528
    0.000


    
Asse risk
889
    0.041
  543.558
   15.521
    0.000

so as per the P value ,we confirm that my data is not normal.but some of the papers I found that they have confirmed normality based on W value.What value should I consider



Based on the above factors can I confirm as my data is not normally distributed .I know the fact that norrmality is a weak assumption.

Thanks in advance
Fadi

[Carlo Lazzaro]

Fadi:
yes, your data are not normally distributed (as it is very often the case outside the statistics textbooks).
If you've tested them before going, say, -regress-, please note that normality is a (weak) requirement for OLS residuals distribution only.
As an aside, please read and act on the FAQ on how to post Stata codes and outcomes on this forum. Thanks.

[Fadi Ansar]

Thankyou so much Carlo Lazzaro .I will take care of the things when I post next time in statalist