Page 1 of 1

IntraDay trading: profit target or market-on-close?

Posted: Mon Apr 08, 2013 7:26 am
by Clivo
Hi - I have an intra-day trading system & I'm trying to identify the optimal balance point between closing at a specific target price R multiple versus running the position until the close & therefore bearing the risk of giving back profit but being in the market to capture tail-type large R multis.

The spreadsheet enclosed details the distribution of multiples for the maximum profit-on-day V profit at close.

Any ideas I'd be grateful for.

Thanks.

Re: IntraDay trading: profit target or market-on-close?

Posted: Mon Apr 08, 2013 8:58 am
by rajivm
Clivo wrote:Hi - I have an intra-day trading system & I'm trying to identify the optimal balance point between closing at a specific target price R multiple versus running the position until the close & therefore bearing the risk of giving back profit but being in the market to capture tail-type large R multis.

The spreadsheet enclosed details the distribution of multiples for the maximum profit-on-day V profit at close.

Any ideas I'd be grateful for.

Thanks.
I always use brute Force method. Just run through the parameters and look at performance.
Thats the advantage we have now with softwares like TB and Fast computers :D

Posted: Mon Apr 08, 2013 5:52 pm
by babelproofreader
I always use brute Force method. Just run through the parameters and look at performance.
Sorry, but I disagree with this. It demonstrates two serious biases: data mining bias and hindsight bias. Data mining bias because, given a set of data, you torture it enough to give results that are pleasing to you; hindsight bias because you are acting on a complete set of data that in reality would not have been available to you, i.e. decisions as to what to do in the earlier part of the test period are informed by data from the yet to occur, latter part of the test period.

My advice to the op is this: your profit taking exit should be determined by your expectation at the time you enter the market - if the entry day is determined to be a trend day with the close likely to be near the high/low of the day then you should hold until the close, perhaps with a trailing stop - if this is not the case then your exit should be placed at that point at which it is expected that there will be no more favourable movement in the position. If you are unable to actually distinguish between these two alternatives then, frankly, you need to do more work on your basic methodology.

Posted: Mon Apr 08, 2013 10:46 pm
by rajivm
babelproofreader wrote:
I always use brute Force method. Just run through the parameters and look at performance.
Sorry, but I disagree with this. It demonstrates two serious biases: data mining bias and hindsight bias. Data mining bias because, given a set of data, you torture it enough to give results that are pleasing to you; hindsight bias because you are acting on a complete set of data that in reality would not have been available to you, i.e. decisions as to what to do in the earlier part of the test period are informed by data from the yet to occur, latter part of the test period.
.
I thought that those were precisely the reasons we backtest. :D
I think someone said History repeats itself .
If Hitler had read about the misadventure of Napolean in Russia, Maybe we would be living in alternate universe now. :D

IntraDay trading: profit target or market-on-close?

Posted: Tue Apr 09, 2013 5:35 am
by Clivo
thank you gentlemen for your food-for-thought.

I am continuing this discussion on the Intra Day forum where I now think it is more appropriate. I hope you will view it there.

Cheers :)