Quote from slapshot:
Frost, I have been reading this thread with interest. Are you interested in some feedback?
IMO, you are trying to be a mechanical automated system trader and yet you are jumping in/out of the market, using "discretion" trying to time the better days of trading to avoid taking losses.
This is not automated, mechanical trading - it is "systemessionary" trading (1/2 system + 1/2 discretionary = messy)
I see this as a veiled attempt to overoptimise. Are you sure you are being honest with yourself about this? Read what you wrote above - you don't seem very committed to your automated strategy. It's obvious you don't trust it enough to trade it every day. You have a lot of "opinions" about which days to trade it - this is not very automated.
Remember that big first day you had? By jumping in/out like you are, you are pretty much sure to miss the next big-ass day for your system that would make up for all of the small losers lately and add big profit to your stats.
IMO, you should trade it every day for all of the trades the system wants to take - or don't trade it at all until you have made it more robust - to where you will be confident doing so.
Someone correct me if I am wrong here.
I am trying to be constructive, not condescending.
Good luck,
Paul
I also have to agree here. Our system was developed with two traders and a programmer. The system does have periods of stagnation (4 or 5 days of small winners and losers) followed by large gains. This is not uncommon especially in certain market conditions.
For example: last month, about half of the gains were from two specific days of trading. Granted, the first day of the month had a large winner, however had we pulled the plug or played around with which days to run we could easily have missed them and only made 50% of the profits.
I get e-mails from the programmer talking about how difficult it is to stomach even the small losses when there is a streak of them. This is completely understandable and in fact expected. To me, these losses are water off a ducks back because I know from my trading experience that these things happen and you get used to them. You must become numb to the losses, confident in the fact that the next day has a higher chance of victory.
I can't recall discussion on the specifics of your trading experience, but I would say if you have a history (years) of manual trading you should apply those same principles of mental resilience - that a trader must posess to survive - to your automated strategy.
If you don't have enough manual trading experience it will be difficult. However, your backtesting seems (on the face of it) to be fairly accurate, so unless you are starting to seriously punch through those drawdown periods (either in length or capital) you just have to weather the storm.
In all honesty if your system does not have the capability to deal with low-volatility trading, developing even a rudimentary filter should be a high priority.
These words are all predicated on the fact that you have faith in the system and that the backtesting is accurate. Only you can really answer those questions.
read either pdfs iv posted on recent pages and say that again. and remember those are both from 'introductory' machine learning courses. basically if you want to get a high paying job researching/programming ai on wall street. you need to know at least that + some. that being said. iv seen job postings for phds in stat arb that were starting salary of +8,000,000$. it is definately not a way to keep 'non mathies' out.
