optimized backtesting
Just read this and am confused as to what it means:
"Structured Mean Variance Optimized portfolios and optimized their back-tests."
... how do you optimize a back-test? According to investopedia, backtesting is "doING a simulation of his or her trading strategy on relevant past data in order to gauge its effectiveness"
So for simplicity, you have 10 arrays, (each a closing price of a different stock), and you have a year's worth of data. Are you implementing a set of rules (strategies) and checking what would happen over the course of that year had you acted on the strategy. Or are you taking that year of data points, extrapolating a reasonable GARCH-like model and running some sort of MCMC which generates data as prescribed by your GARCH-dgp and then checking what would have happened if you acted on your strategies?
furthermore, when someone with basic math background says they "optimized a portfolio backtest" does it mean they entered some parameters into a GUI?
I dont really know what you are talking about with some of the things you say, but optimizing when it comes to backtesting trading strategies is finding a general strategy, and then finding the best parameters in the strategy for your historical data. Ie let's say you have a strategy where you buy a stock and hold it for a day if a stock has had an X% up move the day before. Optimizing means that you find the most profitable X% from your historical data.
We could be talking about somethign totally different though.
Thanks for replying. I have no work experience in trading; I am coming at this from the perspective of a graduate degree in financial econometrics. Can you explain the actual mechanics of optimizing the sample strategy you set forth?
Would you just write a quick optimization program in matlab or so? are there any sources for common backtesting procedures implementation, or are they generally created on an as-need basis?
Voluptatibus consequatur animi dolores modi commodi repellendus. Laborum quia placeat minima. Consectetur explicabo natus ipsa at asperiores qui quasi.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...