Page 1 of 1
Modeling Serial Correlation
Posted: September 30th, 2013, 5:42 pm
by MicroHedge
I am currently using a Lognormal model for stock prices:dS/S = (Mu)dt+(Sigma)dWNow when I look at monthly log return series (historical return data), I can observe serial correlation (0.1-0.2). In order to obtain the serial correlation I simple find the correlation betwee r(t) and r(t+1) series.How do I incorporate this serial correlation into my model?
Modeling Serial Correlation
Posted: September 30th, 2013, 9:36 pm
by MicroHedge
I am trying to see how serial correlation increases volatility (as compared to a simple LogNormal process with same parameters) when I simulate stock prices.
Modeling Serial Correlation
Posted: October 1st, 2013, 5:42 pm
by MicroHedge
Thank you for taking the time to answer. I was finally able to run a model with serially correlated lognormal returns and the resulting cumulative returns do have much higher volatility as time increases. If anyone wants to know how to do this feel free to PM me.Now on to part two, how well does Fisher Geltner Webb equation work when adjusting the volatility paramter for serial correlation.