80% Accuracy in Predicting Intraday Price Moves?

[conduit]

That is nonsense. I could point out 5-10 venues that offer excellent spreads, low enough commission rates and stable liquidity for retail traders to get engaged in. When it comes to market depth that is an entirely different chapter. But your claim is not correct. In fact LMAX is a more expensive trading route than Interactive Brokers for retail traders. We can have an honest discussion comparing the two if you like as long as we stick to facts.

LMAX

It's the only viable place for retail to trade FX.

 

[SunTrader]

Not available here in the U.S. (and for the most part Australia, Canada, Singapore etc) so many of us have no interest either way.

 

[Low Salubrity Thug]

That is nonsense. I could point out 5-10 venues that offer excellent spreads, low enough commission rates and stable liquidity for retail traders to get engaged in. When it comes to market depth that is an entirely different chapter. But your claim is not correct. In fact LMAX is a more expensive trading route than Interactive Brokers for retail traders. We can have an honest discussion comparing the two if you like as long as we stick to facts.

Refering back to the original post:

Interesting Stanford paper:

http://cs229.stanford.edu/proj2011/BrownMundkowskyShiu -PredictingIntradayPriceMovementsInTheForeignExchangeMarket.pdf

Basically, the authors backtested a forex strategy using only the bid and ask sizes at 1 second intervals and achieved a very high predictive ability. They claim a profit of $2800 per day over the 45 period of the test after transaction costs and slippage.

I'm wondering, however, where a trader would get forex market depth, since it's a decentralized market. Do any brokers offer this by API?

I don't know about forex, but this probably won't work on stocks since "level 2" is no longer accurate given HFT, dark pools and all the tricks MMs use to hide orders.

As far as I know LMAX is the only FX execution venue (available to retail clients) which publishes a central limit order book. Happy to be corrected though if I am wrong. Do IB offer this?

It's also pretty essential if trading on the timescales described in the paper that your LP doesn't have the option of 'last look' rejection. If they do then you'd rarely get a fill unless price is likely to move against you. I know someone who's involved in coding MM algos (including FX) and he tells me they use models, not dissimilar to the one described, to predict very short term price movement. They then use these to determine whether to accept or reject an order.

Sure, if you are trading FX on a longer timescale there are a myriad of choices available, but I'm not aware of any venue other than LMAX where the strategy would be viable for a retail client.