Show HN: How I Used Machine Learning to Optimize My Trading Algorithm

Re:Show HN: How I Used Machine Learning to Optimize My Trad... by nolite on Tuesday March 26, 2013 @04:34PM (#5445896)

sticking around for 97% drawdown? Sorry.. doing this in real life would be more stupid than anything

[ Parent | on HN ]

Re:Show HN: How I Used Machine Learning to Optimize My Trad... by unreal37 on Tuesday March 26, 2013 @04:35PM (#5445902)

One point in the graph shows a -144% return. How do you even get a -144% return? Debt?

[ Parent | on HN ]

Re:Show HN: How I Used Machine Learning to Optimize My Trad... by dunster on Tuesday March 26, 2013 @04:41PM (#5445938)

Yeah. This algo is highly leveraged - like 15X. It's possible to really lose your shirt if you trade this algo exactly.

Taibo's algo is interesting as a starting point. It's not one that that you just take off the shelf and start trading with. But, you can take it and learn from it and develop an alternative strategy. Presumably one with less risk!

[ Parent | on HN ]

Re:Show HN: How I Used Machine Learning to Optimize My Trad... by meson2k on Wednesday March 27, 2013 @01:16AM (#5447921)

Dunster, any clue if Quantopian supports option trading?

[ Parent | on HN ]

Re:Show HN: How I Used Machine Learning to Optimize My Trad... by KMag on Tuesday March 26, 2013 @05:54PM (#5446315)

Clearly the simulation doesn't take into account forced de-leveraging at significant losses due to margin calls.

[ Parent | on HN ]

Re:Show HN: How I Used Machine Learning to Optimize My Trad... by loganfrederick on Tuesday March 26, 2013 @04:39PM (#5445926)

Shorting is also a possibility.

[ Parent | on HN ]

Re:Show HN: How I Used Machine Learning to Optimize My Trad... by xntrk on Tuesday March 26, 2013 @05:22PM (#5446158)

yeah the algo looked like it was just highly leveraged version of the benchmark. The benchmark, which you can get a better look at if you remove the algos chart looks a lot like SPX or some other US index (at least when you eyeball it). I would like to see this run against 2008s market and see what would happen.

[ Parent | on HN ]

Re:Show HN: How I Used Machine Learning to Optimize My Trad... by dunster on Tuesday March 26, 2013 @05:26PM (#5446175)

The benchmark on Quantopian is indeed modeled after the S&P 500.

How would the algo do in 2008? It's trivial for you to check it yourself. Click the "clone algo" button, change the time range of the test, and click "Run Backtest." Question answered!

[ Parent | on HN ]

Re:Show HN: How I Used Machine Learning to Optimize My Trad... by Domenic_S on Wednesday March 27, 2013 @03:05AM (#5448221)

> I would like to see this run against 2008s market and see what would happen.

Disaster. I'm only in Nov and at -200% return, 88% drawdown. Ouch.

Edit: spoke too soon: -453% vs -45.3% bench in nov.

[ Parent | on HN ]

Re:Show HN: How I Used Machine Learning to Optimize My Trad... by Hansi on Tuesday March 26, 2013 @05:52PM (#5446308)

Agreed, this is just silly.

[ Parent | on HN ]