Machine designed strategies. Do they work?

[trade4ever2day]

Quote from amazingIndustry:

he never worked with ML. You really should do a bit more diligent homework. ML initially funded the venture with 30% and later on sold its stake in two tranches to BBG. Plus ML was one of the first USERS of the terminals, nothing more nothing less. Please get your facts straight.

Why are you being so obnoxious? These are the facts I posted and you now use them to tell me I am not aware of them? You must be kidding fellow.

If Bloomberg thought like you and was happy to make $10 million a year using his knowledge he would have never build a very useful tool for the whole industry. So you are an empirical idiot when you make such assertions because you generalize and there is at least a sound example that falsifies your general rule.

Keep your tone down please. Because you failed using machine learning it does not mean everybody else will. You think highly of yourself and probably you should not. Your tone exhibits a lot of frustration. Bye-

 

[trade4ever2day]

Quote from braincell:

As for the poster "milewski05" i got in touch with him via e-mail and checked him out. He's a real person and a real TSL user, i know that for sure.

I totally agree with the rest you wrote about TSL but not about this one since I know of no way of checking someone through email. He may be a real user or he may not be, or he may be a user who is doing favors for some reason, I have no idea, how do you?

 

[braincell]

Quote from alexandermerwe:

I totally agree with the rest you wrote about TSL but not about this one since I know of no way of checking someone through email. He may be a real user or he may not be, or he may be a user who is doing favors for some reason, I have no idea, how do you?

Look i just know, i can't share that info since it's private (not just the email), but trust me he's real. I was the first person who thought he was bogus, he's not.

 

[amazingIndustry]

there is tons of evidence that debunks machine learning applied to financial model testing and validation in the same way than neural networks were debunked (even a Google search reveals that and if you really ever worked as part of a quant trading group for either a hedge fund or sell-side desk you would know it first-hand as well). No serious hedge funds uses any of those algorithms especially not Renaissance Tech which I know first hand because some alumni from grad school work(ed) there.

You are the one full of yourself claiming you know about the industry while your posts strongly indicate you are a wanna be trader trained programmer.

P.S. You were utterly wrong M. Bloomberg and his association with ML and now you pretend you did not say what you said? Get outa here...

Quote from alexandermerwe:

Why are you being so obnoxious? These are the facts I posted and you now use them to tell me I am not aware of them? You must be kidding fellow.

If Bloomberg thought like you and was happy to make $10 million a year using his knowledge he would have never build a very useful tool for the whole industry. So you are an empirical idiot when you make such assertions because you generalize and there is at least a sound example that falsifies your general rule.

Keep your tone down please. Because you failed using machine learning it does not mean everybody else will. You think highly of yourself and probably you should not. Your tone exhibits a lot of frustration. Bye-

 

[gmst]

Quote from amazingIndustry:

there is tons of evidence that debunks machine learning applied to financial model testing and validation in the same way than neural networks were debunked (even a Google search reveals that and if you really ever worked as part of a quant trading group for either a hedge fund or sell-side desk you would know it first-hand as well). No serious hedge funds uses any of those algorithms especially not Renaissance Tech which I know first hand because some alumni from grad school work(ed) there.

You are the one full of yourself claiming you know about the industry while your posts strongly indicate you are a wanna be trader trained programmer.

P.S. You were utterly wrong M. Bloomberg and his association with ML and now you pretend you did not say what you said? Get outa here...

Hey are you sure RenTec doesn't uses any machine learning algos ? Are you positive on that ?

Reason I ask is I saw a job spec from a very large fully systematic hedge fund based in Houston (forgetting its name currently, but if you know about hedge funds you know the one I am talking about). Basically a bunch of mathematics/compsc Phds started it back in the day. Now they claim their office gives a panoramic 180 degree view of Houston city. The job spec specifically mentioned machine learning techniques.

Now, if this hedge fund uses machine learning, then why won't RenTec be also applying it ? In any case, this is a very big and very serious and very successful hedgefund so your assertion that no serious hedgefund uses Machine learning is not plausible.

Disclaimer : Since I haven't worked there I have no way of knowing firsthand how much machine learning they actually employ.

 

[omega_350]

for me i would never completely trust any automated system entirely i suggest use a algorithim to enter with a high probability and trade your exit based on your market knowledge or other factors

 

[alexvnew]

Quote from luisHK:

It seems you're mixing up TSL and PAL here. PAL has been commented by several long term ET users in this thread and others and does offer a free demo. Cost is much lower than TSL although I'm not sure about the usefulness of the software.

PAL is expensive but I played with the demo for a couple of days and this is not machine learning but a scanning program for repeatable price patterns. This approach has some merit to it and it is unrelated to genetic programming which IMHO is useless and should be avoided at all costs.

The thing with PAL is that to find high frequency price patterns with high statistical significance one must run it on a number of computers with different settings and the price for the license that allows this type of operation is high.

 

[braincell]

Having a quick look at some job postings from the past 10 days here:

http://jobs.phds.org/machine-learning-jobs/quantitative-finance

it seems even Goldman Sachs is looking for machine learning experts. Why else would all of them put "Machine Learning Jobs" as one of the tags in the job opening?

The GS ad also has this:

Preferred Qualifications

· Experience with any of the following: Data mining, machine learning, Artificial Intelligence (AI), Human Capital Interaction (HCI), Neural networks, Bioinformatics, Medical imaging, Operational Efficiency, Process review via Data Analysis

see?

 

[amazingIndustry]

they mentioned the 180 degree panoramic view of the skyline and then talked about machine learning? Interesting combination. But maybe after the machine learning thing did not work out the only thing that was left to attract new grads to their firm was the mentioning of the spectacular views.

I dont know which hedge fund you talk about as there are several billion+ AUM funds in the Houston greater area. I only spoke of things I claim I know and shut up about other stuff. I am a quant trader and claim to know a bit about the industry and machine learning is definitely not a tool widely used at all nor do I or most other quants see merits in considering it. The famous example of curve fitting (and thats exactly the stuff machine learning algos spit out) is the correlation between changes in butter prices in Bangladesh and s&p 500 index returns. Try to pitch the idea to a sell-side ibank MD and report back.

Edit: Why dont you do a search on NuclearPhynance maybe you find the gems there, at least its (correction: was) a rather professional forum (before questions such as "what is the cheapest broker to trade nasdaq futures" were allowed...recession is a bitch...)

Quote from gmst:

Hey are you sure RenTec doesn't uses any machine learning algos ? Are you positive on that ?

Reason I ask is I saw a job spec from a very large fully systematic hedge fund based in Houston (forgetting its name currently, but if you know about hedge funds you know the one I am talking about). Basically a bunch of mathematics/compsc Phds started it back in the day. Now they claim their office gives a panoramic 180 degree view of Houston city. The job spec specifically mentioned machine learning techniques.

Now, if this hedge fund uses machine learning, then why won't RenTec be also applying it ? In any case, this is a very big and very serious and very successful hedgefund so your assertion that no serious hedgefund uses Machine learning is not plausible.

Disclaimer : Since I haven't worked there I have no way of knowing firsthand how much machine learning they actually employ.

 

[braincell]

Quote from alexvnew:

This approach has some merit to it and it is unrelated to genetic programming which IMHO is useless and should be avoided at all costs.

Ugh... Reading something like this makes me cringe. You obviously have no idea what "genetic programming" is. It's just a very efficient "search algo", there's no difference in the fact that it's still a "search algo". If they used hill climbing or some older kind of selection algo, it would still be a "search" get it? It's just a "search technique" and a very efficient one at that. PAL might be using GP in that when it sees one price pattern working well it tries to modify it slightly to improve it even further. That saves time and makes the searching quicker. Don't confuse the searching methodology with a hypothesis testing methodology.