st: Monte Carlo Simulation w/ Pareto Dist. on Tails

[grosegre@msu.edu]

Dear Statalist:

As a novice user of Stata, I am attempting to construct a Monte Carlo  
simulation modeling a Home Price Index, while correcting for  
in-frequent and extreme index returns and an AR(3) process.

A previous Monte Carlo simulation was developed in excel: with the  
E(x) and std. dev based upon historical numbers; random returns and  
errors define by a Normal Prob. Distribution; and including an AR(3)  
process in returns. However, it  failed to adequately simulate the  
index, which may be due to failing to account for the tail gain/losses  
greater then assumed with a normal prob. distribution. To adjust based  
on extreme-value theory, I would like to adjust by modeling the tail  
returns,  past the 3rd std. dev., with a Generalized Pareto  
Distribution. Also,  I would like to return the mean and std. dev. of  
the returns and, provided negative index returns, am average count and  
distribution of losses.

Any insights into or resources for the correct stata programing  
commands? I appreciate the guidance!

Greg

Gregory R. Grose
Michigan State University
Undergraduate; Finance and Economics
grosegre@msu.edu



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/