My backtested system consistently beats the S&P500

Quote from StockTrader1985:

My question is, how do I verify that this backtest is legitimate? Are there statistical tests that I can run to ensure that the system's success would also apply to today's markets?
More...

I do not know what you mean by "legitimate".

We might reproduce your simulation results if you fully disclose the method.

Your trading results are approximately the same as results that I obtain using Standard And Poors 500 index tracking stock symbol SPY historic price data and a trend following method.

You might make an allowance for slippage and commissions in your calculations.

There are no tests that guarantee that any trading method will be profitable in the future.

What seems legitimate to me

<img src=http://www.indcjournal.com/archives/legitimate.jpg \img>

might not be legitimate to you.
 
Quote from StockTrader1985:

OK forget I asked that second question

Please help me with the first question if you can. Any of you math dudes out there, are there any statistical tests that can be applied to verify the legitimacy of a backtest?
More...

If you cannot answer this question...
Then you are at the level...
Of a dude that dropped out of freshman stats in the first semester at age 18.

Why should anyone take you seriously?

You should not even be trading your own money.
Pros will just take it away... if you insist on playing.
 
You have to be careful when you use fundamental data for a backtest. A lot of the data gets revised after the fact, and the reported values in the database now, were unknowable at the time.
 
Quote from Hook N. Sinker:


We might reproduce your simulation results if you fully disclose the method.

Your trading results are approximately the same as results that I obtain using Standard And Poors 500 index tracking stock symbol SPY historic price data and a trend following method.

You might make an allowance for slippage and commissions in your calculations.
More...


BINGO!

Nothing stellar here. I'd rather just dollar cost average into Vanguard's VFINX or just buy SPY a couple times per year. The results would be comperable to your results minus all the effort. My time's worth more than that.
 
Quote from GTG:

You have to be careful when you use fundamental data for a backtest. A lot of the data gets revised after the fact, and the reported values in the database now, were unknowable at the time.
More...

Excellent comments.
 
Almost forgot....


Now if you can find a system that can return me 11% per year above and beyond that of the SP 500......


I've got some new clients that are looking for a place to throw some money.












That's what I thought.
 
i'm no quant guru but his results might also improve if he dollar cost averages throughout the year right?

if the system is value based then test it against a value based large cap portfolio like the ishares s&p value etf.
 
Quote from StockTrader1985:

Hey guys,

I usually stick to trading currencies, but a few days ago I decided to run a backtest on a system that's based on value investing. Using objective criteria over the first 100 stocks that fit the criteria, I ran the backtest over every year from 1998 through 2006, annually. Although I have saved the results in an excel spreadsheet, here's a quick breakdown of the results:

1) My system returned an average of 11.98% over almost 9 and half years, while the S&P returned an average of 4.64% over the same period of time

2) My total return over the period is 187.46%, while the S&P returned 52.75%

My question is, how do I verify that this backtest is legitimate? Are there statistical tests that I can run to ensure that the system's success would also apply to today's markets?

And lastly, if I am confident in my system, how do I setup a managed fund so that I can trade other people's money?

Thanks
More...

Everyone's systems backtests better than the S&P. Then they crash in walk-forward real world testing. So what else is new?
 
You must log in or register to reply here.
Back
Top