G-BOT

[Saud 5]

I thought some of the smart people on here might find this interesting:

 

http://www.datatime.eu/public/gbot/

 

It's a public project to test algorithms and fully automated trading strategies with a trading platform (robot) and there's a client for (only) Interactive Brokers TWS or Gateway. It's really for fund managers and large investors, but we can download it free for evaluation purposes with a paper trading account, as they're looking for people to test the platform and concepts. Primarily for ETFs and futures but options seem to be fully integrated into it as well.

 

Anyway, there's built-in backtesting, forward testing, and some really unique features - "extracting profit from the price curve by player superposition, and overlaying multiple "dynamic probabilistic cage" or "order clouds" from which the price curve cannot easily "escape" without being scalped. This is implemented through an overlay of multiple strategic layers, each one working with a multi-agent logic and hedging each other (player superposition)."  

 

Obviously not for the novice or casual investor but maybe some of you guys here would like to take it for a test drive

 

[PaulCao 51]

Hi Saud, 

 

Looks like there's extensive instructions on how to set it up with IB, 

 

Why not set it up with your IB test account and let it rip with 1 million dollars and see where it goes in a month, 

 

Best,

PC

[Saud 5]

I am   Just got the response to my request. I have a feeling it may take me a month just to learn how to use it lol

[Saud 5]

Update...

 

I haven't installed it yet (waiting for the latest update) but I spent quite a bit of time going through all the forum threads at various places where he (Tom, the guy behind the project) elaborated his ideas, gone through the results of testing and had feedback from early users. It's so different from anything else that I think it's useful to have an understanding of the concepts behind it before getting into the nuts and bolts. From what I've seen, any problems the early users and testers seem to be having is because of their failure to grasp the ideas it's based on and trying to force their own ideas and methods onto the architecture.

 

First of all, it's not a scalper in the traditional sense (like a HF algo), it's simply designed to buy low and sell high by (what he calls) "wrapping" the price to capture the range as it moves. There's no place for time, it's based only on price points, which move in a sort of channel (but can move to the next channel if the price moves up or down). He rejects both trending systems (zero-sum game) and predictive indicators. Basically, the system just grinds away until the realized profit for each trade reaches the unrealized profit (as he sees it, the main problem  with HFT is that the realized profit can exceed the unrealized) and then it ends that trade. So drawdowns during the course of that trade are not thought of as drawdowns, but as "investments", since the system will increase the trade size on an up or down trend (think mean reversion, in terms of the price wrapping). It's meant to run 24/7 through Gateway.

 

He's pretty coy about what's in the black box but the whole idea is that the system "supervises" a number of virtual "players" who are bidding against each other to create what's called an "order cloud" that's superimposed on the points of the price wraps (and there can be multiple overlays). The players are trend buy, trend sell, countertrend buy, and countertrend sell, and their activities determine the buy/sell. The main strategy is sort of a superposition of two trending strategies which seem (long-term) like a countertrend strategy. I admit I don't understand that bit just yet.

 

The hedging aspect of it comes from being able to run multiple instances of strategies on the same instrument (or folio) using "clones" or even multiple bots on the same account. This also keeps each position size quite small and the portfolio "massively" diversified overall. Of interest to us here is the way he's integrated options into it. (This in only the most recent iteration of the system so it probably needs refinement). The bot uses option "corridors" to hedge against the runaways, so usually there will be

 

+2 calls

- call

 

instrument

 

+2 puts

-1 put

 

One long is to hedge, the other long covers the short, the short pays for the longs. The short credits must equal or exceed the cost of the longs, so their strikes will be near or ATM and the longs further out, with expiries a few months away. It's a semi-automatic process, there's a little "retrieve options" function that populates with the appropriate ones to use. Other combinations like +3/-2 could also be used.

 

Now, the whole idea (since it's targeted at hedge fund managers and large investors) is that the user is freed up from the minutiae and can see the larger strategic picture (allocation etc). So there's every level of discretion available from fully automated on down, and it's intended that each user sort of find and refine their own strategies (with the built-in testing engine which he designed to really be a stress test with the most vicious conditions possible).

 

Anyway just some preliminary impressions in case anyone is interested