Has anybody ever had a consistently profitable algorithm automated?

Quote from pmg:

...
I am now using tradestation and have been designing all kinds of simple trading systems in Easylanguage, which are all based on indicators, some very simple, others more complex. I have tried trend following, break-outs and purely indicator based systems.

However, with input optimization it was easy to have positive backtest results, but I never managed to come up with an algorithm, that can consistently show even small profits on more than 1 symbol.
...
More...
At risk of attracting the TA police .. I saw similar results in my own indicator backtesting a number of years ago.

Nothing useful in the TA world in my view. Just pretty charts that captivate morons & uninformed "investors".

Single instrument historical price is not particularly useful.

Other things can be, depending on your speed, costs & quant work.
 
IMO, any aspiring algo developer has to go through a certain rite of passage. You need to write systems that are based on all sorts of indicators. Experiment with all of them. Combine this one with that, etc, etc.

After you discover after months of tweaking that these things do not work consistently over time, you can move on to systems that do.


I currently have a dozen strategies that are mostly variations of the same strategy. They are consistent. I don't expect them to be consistent forever. However, I will take advantage of them as long as they work (I've been running them for years and haven't seen any sign of performance changing). The strats do look at patterns on different time frames, however, the meat of the work is done by reading the level 2.


The key imo, is to explore all kinds of insane ideas. The crazier, the better. That means less people are looking into it.
 
Interesting, I just read this thread. (my first post on this site!)

I believe it is possible. I'm doing it myself now with real trades. Learnt from someone who has been working on this for a few years now and has programmed systems for gold, oil, russell and s&p.

Did you find any good literature and /or guidance??
 
well, to summarize it seems there are some people, wo are confident about their algos. However I don't really see a pointer in the right direction so far, but that kind of fulfills my expectation.
the folks, who trade profitably, will probably not hang out in message boards.
 
That makes sense.
Though there are some people that actually help out. Ive seen some very helpful comments on this forum! I know there are a few blogs that are sharing their trades with automated trading right now. Its a great learning experience.
 
You must log in or register to reply here.
Back
Top