SAS/IML Software and Matrix Computations

Statistical programming, matrix languages, and more
BookmarkSubscribeRSS Feed
bonfa
Calcite | Level 5

Good afternoon everyone,

I need to test the distribution of the ratio of two normal variable. I believe that this variable follows the Cauchy distribution. Then I would like to know which procedure can test this hypothesis. And, if this variable does not fit in the distribution of Cauchy, What procedure that I can use for evaluate which distribution the variable follows?

Best regards.

1 REPLY 1
Rick_SAS
SAS Super FREQ

I've never done this, but here are some ideas that might help:

1) For a graphical test, use the quantile-quantile plot, as described in this article Modeling the distribution of data? Create a Q-Q plot - The DO Loop

2) Transform your data by the inverse CDF and then test whether the result is uniform.

3) You can use PROC NLMIXED to use MLE as described in this paper for the t-distribution: http://www2.sas.com/proceedings/forum2007/181-2007.pdf

The Cauchy MLE is easy find online.

Of course, if the normal variables are independent, then no need to test, since this is the definition of the Cauchy distribution.

Regarding the question "if this does not fit, what procedure can I use to evaluate which distribution...,"  I recommend that you look at the SEVERITY procedure in SAS/ETS software.  Here is an example: SAS/ETS(R) 13.2 User's Guide

Good luck.

sas-innovate-white.png

Our biggest data and AI event of the year.

Don’t miss the livestream kicking off May 7. It’s free. It’s easy. And it’s the best seat in the house.

Join us virtually with our complimentary SAS Innovate Digital Pass. Watch live or on-demand in multiple languages, with translations available to help you get the most out of every session.

 

Register now!

From The DO Loop
Want more? Visit our blog for more articles like these.
Discussion stats
  • 1 reply
  • 1792 views
  • 0 likes
  • 2 in conversation