47157_en-US

liliconcom

goodness of fit - include in a script

Dear all, 

 

I'm triyng to add a goodness of fit after a distribution, using a script. There is a loop. The analysis is performed for n columns.

 

this is the part of the code:

 
 disti = dt << Distribution(
  stack( 0 ),
  Continuous Distribution( 
  Column( listName[i] )),
  Normal Quantile Plot( 1 ),
  Fit Distribution( Normal(Goodness of Fit( 1 ) )),
  Capability Analysis( 1 ),
  Invisible( 1 ) 
   
 );
 

But no goodness of fit appears... and  I don't konw why..

Do you have an idea?

 What does it mean the "1" parameter of this function?

 

Thank you 

 

Maïna

Re: goodness of fit - include in a script

yes, all my columns are in continuous type 

Re: goodness of fit - include in a script

Are the columns you are analyzing of Modeling Type Continuous?  If they are Ordinal or Nominal, they will not provide a goodness of fit

Re: goodness of fit - include in a script

Or do I have no place to display the result?
Would that be possible?

Re: goodness of fit - include in a script

I used your script, and just attached a different data table, so that is why I suggested that you need to look elsewhere.

Re: goodness of fit - include in a scriptNo, It works well !Re: goodness of fit - include in a script

Do you have the same problem if you run Jim's script?

Re: goodness of fit - include in a script

Thank you Jim, 

I have already do this loop IThe script is bigger ! 

My first script was like that:

 

The distribution appears, the capability analysis too, but not the goodness of fit test... 

I don't understand why ;(

 

 disti <SPAN class="token operator">=</SPAN> dt <SPAN class="token operator"><</SPAN><SPAN class="token operator"><</SPAN> Distribution<SPAN class="token punctuation">(</SPAN>
  stack<SPAN class="token punctuation">(</SPAN> <SPAN class="token number">0</SPAN> <SPAN class="token punctuation">)</SPAN><SPAN class="token punctuation">,</SPAN>
  Continuous Distribution<SPAN class="token punctuation">(</SPAN> <SPAN class="token function">Column</SPAN><SPAN class="token punctuation">(</SPAN> listName<SPAN class="token punctuation">[</SPAN>i<SPAN class="token punctuation">]</SPAN> <SPAN class="token punctuation">)</SPAN> <SPAN class="token punctuation">)</SPAN><SPAN class="token punctuation">,</SPAN>
  <SPAN class="token function">Normal Quantile</SPAN> Plot<SPAN class="token punctuation">(</SPAN> <SPAN class="token number">1</SPAN> <SPAN class="token punctuation">)</SPAN><SPAN class="token punctuation">,</SPAN>
  Capability Analysis<SPAN class="token punctuation">(</SPAN> <SPAN class="token number">1</SPAN> <SPAN class="token punctuation">)</SPAN><SPAN class="token punctuation">,</SPAN>
  Invisible<SPAN class="token punctuation">(</SPAN> <SPAN class="token number">1</SPAN> <SPAN class="token punctuation">)</SPAN>

 But I want to add a goodness of fit analysis below the histogram of the distribution.

Tks 

Re: goodness of fit - include in a script

I have expanded your script slightly, to make it a functioning script.  Your code works well using the Semiconductor Capability data table.  I suggest that you look at the makeup of your data or the list, "listName".

Names Default To Here( 1 );
dt = Open( "$SAMPLE_DATA\semiconductor capability.jmp" );

listName = dt << get column names( numeric, continuous );

For( i = 1, i <= 2, i++,
 disti = dt << Distribution(
  stack( 0 ),
  Continuous Distribution( Column( listName[i] ) ),
  Normal Quantile Plot( 1 ),
  Fit Distribution( Normal( Goodness of Fit( 1 ) ) ),
  Capability Analysis( 1 ),
  Invisible( 1 ) 
   
 )
);