Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: montecarlo simulations with actual distributions


From   "Austin Nichols" <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: montecarlo simulations with actual distributions
Date   Wed, 2 Jul 2008 13:13:30 -0400

Emanuele Canegrati:
The mean and sd are unlikely to fully characterize the pdf of returns
(there is serial correlation, heterogeneous kurtosis, and stochastic
volatility to worry about, at least).  I think you might want a block
bootstrap, i.e. your generation of random datasets will select groups
of contiguous obs from the actual data.  Simulations are tricky at
best, but simulating stock market returns is a real bear (or bull,
depending), so I would advise you to get someone experienced with
simulating stock market returns to become a coauthor and help you out.

On Wed, Jul 2, 2008 at 6:30 AM, emanuele canegrati
<emanuele.canegrati@hotmail.it> wrote:
> I have to perform some montecarlo simulations of market returns with STATA. I would like to generate these returns from the original time series under the condition they reflect the actual PDF and so I think I need to calculate the mean and the standard deviations of returns. Which is the programme I have to write? I have to perform any non-parametric analysis before in order to calculate the actual distribution of my returns?
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index