Simulation Procedures
Instructions:
- These procedures can be downloaded
individually, or together as one zip file: sim.zip . Place all files (including
declaration files ending in ".dec" into your
"/gauss/src" directory.
Individual Programs:
- estwhite.g & estwhite.dec Procedure with
necessary declaration file to estimate an
Arma(1,1)-Garch(1,1) model with possibly t-distributed
errors and/or extract the whitened errors which can be
used for bootstrapping functions of interest such as a
terminal stock price distribution. (Note: requires
whiten.g & whiten.dec procedure below to work)
- whiten.g & whiten.dec
Procedure with necessary declaration file used to whiten
data which follows an ARMA(1,1)-GARCH(1,1) Often called
within estwhite.g above.
- boot.g &boot.dec
Procedure with necessary declaration file, used to recolor draws of
whitened error terms according to an ARMA(1,1)-GARCH(1,1)
process. Useful for simulating terminal asset prices in
conjunction with estwhite.g above.
- simgarch.g & simgarch.dec
Procedure with necessary declaration file used to
simulate an Arma(1,1)-Garch(1,1) model with possibly
t-distributed errors.
- simjdiff.g
Procedure used to
simulate a jump diffusion process with stochastic volatility allowing
for possible correlation between asset returns and the volatility
process. Theoretical moments for the asset price at some date in the
future (given the parameters of the process) are also give. Note: This is
the process used in Bates (1996), "Jumps and Stochastic Volatility:
Exchange Rate Processes Implicit in Deutsche Mark Options." Review of Financial
Studies, vol 9, no 1, 69-107.
- rndgams.g
Procedure to generate a matrix of random gamma variates.
[Cameron
Rookley's Home Page|
Gauss to Matlab Resources]
Last modified: 1/1/2003