Pareto density function?

[#1]

I'm working with some inventory data that needs to be evaluated with a pareto.   So basically rank the inventory quantities in order from 1 - yada yada and then fit the pareto.  I know we have the pareto function in the reliability thing but I can't seem to extract the parameters from that.  I have to do this for like a few hundred distributions so what I would love is something like the distribution tool but specific for pareto.  

So largest volume item in position 1, next volume in position 2, yada yada.

I do have a way to sort of make it work in a distribution platform where you set the inventory size as Y/column and the Freq  (so I have 6k units where the batch size is between 2-3k), which gives me something I can sort of use, but its not ideal.

Anyways, any ideas on how to extract the pareto parameters?

[#2]

What about this?

Names Default To Here( 1 );
dt = Open( "$SAMPLE_DATA/Quality Control/Failure Raw Data.jmp" );

obj = dt << Pareto Plot( Cause( :failure ), , Per Unit Rates( 1 ));

report(obj)[Table Box( 1 )]<< make into data table;

[#3]

Was really looking for estimator of the alpha parameter for the PDF/CDF. 

[#4 · Accepted solution]

still not really sure what you need, and if you need a scripting solution etc.

If you're looking at the example script and screenshot,

do you want the fit parameters of the distribution, like for the exponential fit below?

This would be no problem, you could JMP let calculate, in that case you save the formula and further process if needed ...

Names Default To Here( 1 );
dt = Open( "$SAMPLE_DATA/Quality Control/Failure Raw Data.jmp" );

/*
obj = dt << Pareto Plot( Cause( :failure ), , Per Unit Rates( 1 ));
report(obj)[Table Box( 1 )]<< make into data table;
*/

dt_sum = dt << Summary( Group( :failure ), Freq( "None" ), Weight( "None" ) );
dt_sum << Distribution(
	Continuous Distribution(
		Column( :N Rows ),
		CDF Plot( 1 ),
		Fit Normal( Show Fit( 0 ) ),
		Fit Cauchy( Show Fit( 0 ) ),
		Fit Student's t( Show Fit( 0 ) ),
		Fit Lognormal( Show Fit( 0 ) ),
		Fit Exponential,
		Fit Gamma( Show Fit( 0 ) ),
		Fit Johnson( Show Fit( 0 ) ),
		Fit SHASH( Show Fit( 0 ) ),
		Fit Normal 2 Mixture( Show Fit( 0 ) ),
		Fit Weibull( Show Fit( 0 ) )
	),
	SendToReport(
		Dispatch( {"N Rows"}, "Compare Distributions", OutlineBox, {Close( 1 )} ),
		Dispatch( {"N Rows"}, "Quantiles", OutlineBox, {Close( 1 )} ),
		Dispatch( {"N Rows"}, "Summary Statistics", OutlineBox, {Close( 1 )} ),
		Dispatch( {"N Rows"}, "Fitted Exponential Distribution", OutlineBox, {Close( 1 )} )
	)
);
		

[#5]

So all of those density functions you generated there provide parameters for them.  So does the Pareto distribution, which is a special case of the exponential distribution.  I was just trying to think through how to generate the alpha parameter for the pareto distribution using the Pareto platform in the same way the distribution platform will do that for the other distributions (like the fitted exponential distribution on the bottom of your report). 

That said, I think your script showed me how I can set up the data so that I can model it with an exponential directly (which is perfect).  What had messed me up was that my data was in the un-summarized form so I couldn't get that distribution to work correctly.    

Thanks much!