Friday, 20 August 2010

Progress in Jonathan-Jerome trading

Over the week the detailed process of tightening up the trading process has occured and Jerome is perhaps 95% there already. He has done a major sort out of his laptop and now has it hooked up to two widescreen TVs in his flat. Apparently the clean up has resulted in a massive improvement in the laptop's performance so he is very pleased with this.

As I begin to do my final work on my LSE dissertation, Jerome and I will be attempting to ensure that we get exactly the same results from the trading systems. I have programmed the systems onto Jerome's laptop and it is now just a matter of interpretation. One outcome of this was that I produced a one-page schedule in which the systems are specified exactly. This will now form the basis of what we do going forward. We will have two weeks of running this on and off and then September 2nd and 3rd will be real-time trading tests, with the week commencing September 6th when we still plan to go live.

That said, this week hasn't been a great trading week, though we have still managed a small profit. Three losing days in five, which is quite a lot worse that usual, but not impossible.

I have two pending research projects. Firstly, I am looking closely at trade execution and the issue of whether bad fills could have a material impact on us. So far the result is that this shouldn't be a problem. Even if every trade went through at a price that was 1pt worse than expected we would still make 50% of the theoretical profit.

Secondly, I have been looking at some slower intra-day and two or three day systems which we might blend in at some point in the future. These have lower Sharpe Ratios that the high-frequency trading but also have a higher average trade profit, which makes them less sensitive to bad fills. These might be worth looking at closely, but I also want to avoid over complicating the trading from the start.

So everything looks very promising at the moment

No comments: