I've programmed a backtesting framework where I'm trying models / strategies / algorithms over historical data. What passes validation goes into a live trading system (very minimal but functional). Main problem: well, no killer strategy yet. Although I'm starting to get the feeling it's more like having lots of them monitoring the market for appropriate conditions. So far I've been working alone, after the day job. Time is one problem (I've a limited amount of it in this setup), sharing ideas and getting another point of view is another. Therefore I'm thinking of hiring / associating myself with another quant developer. I'm keeping my job, but have got some $50k of capital. If an associate doubles it, we establish an LLC and hire someone, in Europe this should comfortably buy at least 2 years of development. This would solve the time problem (the hired employee would work while I/we cannot) and the knowledge problem: it'd be three of us to tackle the problem of beating the market. I wonder if: 1) You think this might work. 2) Anyone's interested (please P.M. me if so). Of course you realize interest doesn't necessarily lead to a deal.
For the vast majority of options strategies, you need portfolio margin and that starts at 110k. Two points: (1) Instead of searching for a single "killer strategy", you would almost certainly be better served by developing a collection of simple de-correlated strategies (think lower Sharpe, but high t-stat) (1) don't concentrate on options only - there are other assets at your finger tips
Classic entrepreneur mistake is to scale the business (hire, etc) without a business model. Backtest, trade small and manually, find some simple and small wining strategies. If you can, use technology to reduce your manual input and only then you scale. You get someone to follow those manual steps you already know work. Otherwise you’ll burn our capital holding a big team. You’re more agile by yourself.
>> Classic entrepreneur mistake is to scale the business (hire, etc) without a business model. Backtest, trade small and manually, find some simple and small wining strategies. Oh, no. "Backtest, trade small and manually, find some simple and small wining strategies" IS THE BUSINESS MODEL
But you want to put your 50k (x2 with investor) to a developer... and then what? Does it actually work? Have you got more money to put down for the trading side of it? Because if you think you can get lots of money in a fund-style deal... I doubt you'll get any... you don't have any track record of statistical significance (real trading).
>> But you want to put your 50k (x2 with investor) to a developer... and then what? No way a developer. Plain developers are useless for what I need now (I'm a developer myself so programming is the least concern). >> Does it actually work? I guess I'll find out. >> Have you got more money to put down for the trading side of it? I have my own capital, although not enough to live off it. >> Because if you think you can get lots of money in a fund-style deal... I doubt you'll get any... you don't have any track record of statistical significance (real trading). Surprisingly, I've got a lot of floating capital around me (sums from $10k to a couple millions) and nope, it doesn't take years of trading to access it. But it does take real trading and working, well risk-managed strategies. Which I don't have yet. I have real "discretionary" trading, exasperated that the systematic stuff still doesn't show enough potential. And lost So I'm better getting back to systematic and trust what the system tells me (afterall I thought and programmed it).
Think it like this: everyone (here and outside this forum) explains me how I should do this business *once I come up with a working, profitable, proven, risk-managed, live-tested, years-long in the market strategy*. Like I freakin' don't know that. Also everyone in this business is a highly successful, very profitable, proven, risk-managed, live-tested, years-long in the market trader with dozens of strategies and constantly coming up with new ones. Only they can't tell me what they are and not even show me their trades. I'd take my chance at reverse engineering them, obviously, but at least I could run them through some sanity checks to test those claims. No such thing available, though. So tell me then, before your golden eggs, how did you prepare the chicken? Did you all off a sudden pulled those things out of your ****? Or it was more like a pseudo-deterministic process with years of learning, experimenting, taking informed guesses, failing, figuring out why and what's wrong, rinse and repeat with the feeling that you're getting closer? Coze the latter is my approach and I'm trying to hasten it up by partnering with someone who brings something similar to the table. But if I'm doing this all wrong and pulling the egg out of my **** works, please tell me what should I change in my diet.
In terms of trader's knowledge I'm neither a beginner, so please stop the condescendence, nor at the end of my career, thus rotten rich. As a developer, I'm in the top 5%. I could go 2% but if you're a trader and know probability even intuitively you realize that's not "just" a 3% effort. It's going from covering 20 cases of the fucked up universe that developers encounter to 50 cases. 1/20 to 1/50. It took me 15 years to cover 20 cases so figure out how much time I still need to double that *and* add 50%. For real. Because if I'm to trick the system (of interviews), I can go from 5% to 1% in a year. Did that once in my life when it was really, really necessary, got from rejected to admitted first. But now, I AM THE SYSTEM. I won't steal my own hat, if I'm to succeed it's gotta be for real.