Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Lessons learned building an ML trading system

tradientblog.com

Post reply on HN
Print view View original

141–150 of 160 posts

Re: Lessons learned building an ML trading system

#141

Post by smabie »

traK6Dcm wrote: post #74
smabie wrote: post #71

Earlier quoted context omitted.

How do you maintain a market neutral portfolio without the ability to go short? Also I would be interested in hearing about your Sharpe ratio. Thanks! Great article!

You can't go short but you can still hedge your exposure by going on short on something very correlated - like Bitcoin futures. I do this manually depending on how much capital I'm trading. I actually have no idea what my Sharpe ratio is, sorry. With the system constantly changing, data formats changing, exchange balance apis changing, and accurate monitoring already being a challenge in itself, it's very difficult t…

If you don’t know your Sharpe, how do you know you are actually making alpha? Without knowing your return and volatility, you can’t really evaluate the performance of your algo or know if changes are helping.

Re: Lessons learned building an ML trading system

#142

Post by TomMarius »

mrpopo wrote: post #109
traK6Dcm wrote: post #61

Earlier quoted context omitted.

Author here. Actually, the system is mostly taking liquidity from the market so it's not even doing that :) Perhaps, "opportunity for others to create more liquidity" Jokes aside, it's actually something I am thinking about a lot. Such systems don't create value, but they extremely intellectually interesting and I've learned a lot. You can say the same for many other projects, for example academic research in many fi…

> But it's intellectually interesting, so people work on it. No. It's a financially sound way to use time, so people work on it. Anything can be intellectually interesting. What kind of academic research does not create value? And if so, maybe it deserves to be criticized the same way.

> What kind of academic research does not create value? And if so, maybe it deserves to be criticized the same way.

I have friends that study social behavior of ants. Is there any way to apply that?

Re: Lessons learned building an ML trading system

#143

Post by deepnotderp »

jacquesm wrote: post #90
thundergolfer wrote: post #76

This post is an exemplar of the crucial relationship between domain-specific knowledge and ML competency in the ML space. The bulk of the post is detailing the tricky ins and outs of trading, and overall the author gives the impression that they're broadly knowledgeable about stock markets. Contrast this post with those you see with ML hobbyists who delve into medicine or fake-news and produce useless results testame…

Domain knowledge is essential to almost any project that aims for eventual commercial success, it is quite rare than an outsider will come into a field, apply some ML and make a killing.

RenTec

(Yes, I know they do much more than ML, but still)

Re: Lessons learned building an ML trading system

#144

Post by throwawaymath »

deepnotderp wrote: post #143
jacquesm wrote: post #90

Earlier quoted context omitted.

Domain knowledge is essential to almost any project that aims for eventual commercial success, it is quite rare than an outsider will come into a field, apply some ML and make a killing.

RenTec (Yes, I know they do much more than ML, but still)

That is not what Simons did to make Renaissance Technologies successful. Simons cultivated domain knowledge long before he started a hedge fund, because he had an interest in trading and gambling even as a professor. He also hired people with financial experience.

The historical record overlooks the people he hired who knew a thing or two about trading, while fixating on the team of NLP scientists he hired from IBM. Likewise Simons wasn't initially successful in the very, very early years. It wasn't until the late 80s that the Medallion firm really came into its own.

Re: Lessons learned building an ML trading system

#146

Post by jotakami »

rixrax wrote: post #25
nickreese wrote: post #11

After having spent an insane amount of time in late 2017/2018 building an HFT bot for Binance I can say this is a pretty solid article. In our case we were doing triangle trading between BTC/ETH/USDT pairs and had our buys/sell delay down to 3-7ms. At one point moving 0.3-0.7% of Binance’s daily volume. Few notes: * Finding an objective point of truth for value when all of the currencies are floating is hard but vita…

When You do this kind of (HFT) trading, how does ones tax returns look like? E.g. I assume you still have to report every trade with its gain/loss to a tax authority?

Depends on the nature of the trades. If you are hedging and maintaining a delta-neutral position (very common situation for market makers and arbitrageurs) then technically your trades are part of a straddle, which is reported separately from capital gains. (in the U.S., at least)

I learned this because I had a six-figure capital loss on my spot crypto trades, and a slightly larger gain on unregulated futures which do not fall under capital gains. Capital losses can only offset up to $3k in other income, so I was terrified that I’d have to pay taxes on the massive “gain” without being able to count the losses against it. Fortunately, the straddle is the correct way to report this, allowing gains and losses to be matched across instruments that may otherwise fall into different income categories.

Re: Lessons learned building an ML trading system

#147

Post by jotakami »

d--b wrote: post #131

Is anyone else suspicious about the results? Claiming a 4000% return while staying market neutral seems a little too good to be true. First: those levels are insanely high, so the algo must be taking some absurd risks and have the worst sharpe ratio, or getting pretty close to being 100% accurate. Second: if you can scale this across markets, and assuming the same return, that investment will turn into 12 billions in…

Back in 2017/2018, not at all surprising. I don’t see that kind of opportunity in the crypto markets anymore though, they’ve gotten a lot more efficient. Used to be able to scalp 1% on big price moves in altcoin futures at least once a week, now the prices move in lockstep with spot.

Re: Lessons learned building an ML trading system

#148

Post by jotakami »

traK6Dcm wrote: post #74
smabie wrote: post #71

Earlier quoted context omitted.

How do you maintain a market neutral portfolio without the ability to go short? Also I would be interested in hearing about your Sharpe ratio. Thanks! Great article!

You can't go short but you can still hedge your exposure by going on short on something very correlated - like Bitcoin futures. I do this manually depending on how much capital I'm trading. I actually have no idea what my Sharpe ratio is, sorry. With the system constantly changing, data formats changing, exchange balance apis changing, and accurate monitoring already being a challenge in itself, it's very difficult t…

Ever try trading futures basis and spreads? I was killing it in mid-2018 in that space, although I got cocky and used my winnings to make foolish leveraged bets that did not end well.

Re: Lessons learned building an ML trading system

#149

Post by perl4ever »

growse wrote: post #126
perl4ever wrote: post #91

Earlier quoted context omitted.

We obviously can have liquidity on in-between timescales, because we had it before modern HFT trading, right? The average person may be overly paranoid about HFT, but it doesn't make sense to say they benefit from it, because they are not going to be in a position where they benefit from an execution in a fraction of a second. Price improvement of fractions of a penny has gotten silly too. It's easy to think of it as…

Before HFT, we had enormous spreads and expensive trades on many instruments in many markets. Smaller spreads lower costs for everyone: institutional, retail etc.

When exactly are you thinking of? Are you old enough to remember before decimalization? Do you consider that "enormous"?

Re: Lessons learned building an ML trading system

#150

Post by deepnotderp »

throwawaymath wrote: post #144
deepnotderp wrote: post #143

Earlier quoted context omitted.

RenTec (Yes, I know they do much more than ML, but still)

That is not what Simons did to make Renaissance Technologies successful. Simons cultivated domain knowledge long before he started a hedge fund, because he had an interest in trading and gambling even as a professor. He also hired people with financial experience. The historical record overlooks the people he hired who knew a thing or two about trading, while fixating on the team of NLP scientists he hired from IBM.…

Yes, I understand that the "financial naiveness" so to speak of RenTec is overplayed in the media, but my point is that superior domain knowledge wasn't what enables Medallion to win. And to be fair, most of the early years were before systematized quant trading.
Post reply on HN