Search results

Usually running a trading bot isn't very performance or memory intensive, so probably any VPS will do. To verify you could check in the task manager when it's running locally how much CPU & memory it's using. It's probably more critical to have your VPS close to the exchange to minimize latency...

Looks like I'll go with polygon.io since it's pretty much the only data provider with aggregate 5sec intraday bars I could find from a long list of data providers. It doesn't seem to have an option to filter out pre-/post-market data though, which would be nice to reduce the download, so now I...

This was with MQL5 for MT5 and yes C++ implementation is couple of orders of magnitudes faster. However, when you increase the number of instruments and optimization interval even C++ implementation takes awhile.

Yes you are right, it's not necessarily a trivial task to port CPU code to an efficient GPU implementation and doesn't automatically mean that running code on GPU is faster. However, I have 20+ years of professional experience writing code for GPUs so I know a thing or two about it ;) I think...

Lol, I'm not optimizing 100 free parameters :D This is just simple matter of fact that even 20k parameter permutations takes long time to optimize.

Even for the most simplest MT5 strategy it takes about 100 mins for single single forex pair to run 10 year walk forward analysis (4 year in-sample window, 1 year out-of-sample WF phase) on 18 core PC.

Interesting, the optimization shouldn't require much memory even with 5sec bar resolution, let alone 1min bars. One year of 5sec forex bar data is only about 200MB and only 17MB with 1min bars.

Distributing the work to cloud services is definitely a good step, but utilizing GPUs can dramatically cut down the computation time and cost even when running on cloud. It's also nice to be able to iterate locally with reasonably priced hardware before running on cloud with a larger data set...

To speed up strategy parameter optimization, I have been considering adding GPU support to my trading platform. I have thought this mainly for strategy development and portfolio generation, but not for live trading nor backtesting which are not as performance intensive. Are people using GPUs to...

Implementing the strategies in terms of trade limits

Depending how you have designed your trading system, you could eliminate slippage caused by latency.

Sticking to the stocks only for now since stock's only thing I can trade atm. Got my hands full getting my trading platform working even with stocks ;)

Overrated by who? If you ask general public with limited understanding of AI, then probably yes because they think AI can do miracles. If you ask people with some understanding of the limitations of AI, probably not and they have better understanding which tasks AI is suitable for. Like with any...

It would be an interesting test, to see what parameter optimization strategy would give the best results.

I give you that TWS nor the server side doesn't exactly appear to be a pinnacle of software engineering. I once went back and forth with an IB support engineer trying to explain some pretty obvious bug in their API/server and all it came down to was "it's a feature not a bug" kind of...

The "order slippage" is the delta between the algo limit price to trigger an order and the streamed market data price. That could be caused by the delay in the stream data, but also gaps in the sales prices. I'm just using the last sale price.

I measure two types of slippages both of which can have significant impact on P&L. One I call "order slippage" which is the slippage of getting the price info to act upon and to submit an order, and the other "fill slippage" which is the delta between the price at the time of the order and the...

I definitely need to improve the order execution and that's on the todo for my trading system. I'm using market orders current to ensure the orders get (eventually) filled. Even when the paper orders are not executed at the exchange (obviously) the fills can be simulated for the better fill...

Yeah, not sure what's going on. I see this occasionally happening when submitting a batch of ~15 orders at the beginning of the day (~9:31 or so) when the market is very volatile. Most of the orders gets filled fast but few may linger for several minutes even when there's volume showing for the...

That's not what I'm asking. I'm asking how realistic the paper trading fills simulated by IBKR are. The market impact of retail trader orders is irrelevant. This can be tested by running the same systematic trader simultaneously on paper and live trading accounts and I'm wondering if someone has...