SERVING THE QUANTITATIVE FINANCE COMMUNITY

 
User avatar
supportoranges
Topic Author
Posts: 9
Joined: February 13th, 2013, 3:38 pm

Sharpe Ratio Based on GA Simulations

July 12th, 2014, 11:45 am

I am optimizing parameters by running simulations, each simulation processing 2012-2013. After each 'run' I come up with a return representing all the trades which the system took. For my GA, I am specifying initial population of 1000 individuals and I find out how each individual performs partly by inspecting how much money was made or lost for each of the 1000 simulations.Is it ok to use the return at the end of a simulation to calculate Sharpe ratio for the simulation, thus each 'Individual' gets a Sharpe ratio? My main concern is that a simulation is for whatever time period I get the data for, and that won't necessarily be a year, it may be 6 months or it may be 5-years.Thank you.
 
User avatar
supportoranges
Topic Author
Posts: 9
Joined: February 13th, 2013, 3:38 pm

Sharpe Ratio Based on GA Simulations

July 12th, 2014, 12:10 pm

To clarify I would like to use all the trades within each simulation to compute my Sharpe. Thus the Sharpe would represent performance of the strategy.
 
User avatar
purbani
Posts: 89
Joined: July 14th, 2002, 3:00 am

Sharpe Ratio Based on GA Simulations

July 12th, 2014, 5:21 pm

The convention is to annualise so you can show your nice results to other people when looking to raise capital - but as long as all the simulations are for the same time period it is still an apples and apples comparison.
 
User avatar
supportoranges
Topic Author
Posts: 9
Joined: February 13th, 2013, 3:38 pm

Sharpe Ratio Based on GA Simulations

July 12th, 2014, 8:22 pm

I am so thankful for your response and it will help me to move on. Thanks again!
ABOUT WILMOTT

PW by JB

Wilmott.com has been "Serving the Quantitative Finance Community" since 2001. Continued...


Twitter LinkedIn Instagram

JOBS BOARD

JOBS BOARD

Looking for a quant job, risk, algo trading,...? Browse jobs here...


GZIP: On