Voodoo Spectrum of Machine Learning and Data Sets
649 Views
I used to be very gung-ho about machine learning approaches to trading but I’m less so now. You have to understand that that there is a spectrum of alpha sources, from very specific structured arbitrage opportunities -> to stat arb -> to just voodoo nonsense.
As history goes on, hedge funds and other large players are absorbing the alpha from left to right. Having squeezed the pure arbs (ADR vs underlying, ETF vs components, mergers, currency triangles, etc) they then became hungry again and moved to stat arb (momentum, correlated pairs, regression analysis, news sentiment, etc). But now even the big stat arb strategies are running dry so people go further, chasing mirages (nonlinear regression, causality inference in large data sets, etc).
In modeling the market, it’s best to start with as much structure as possible before moving on to more amorphous statistical strategies. If you have to use statistical machine learning, encode as much trading domain knowledge as possible with specific distance/neighborhood metrics, linearity, variable importance weightings, hierarchy, low-dimensional factors, etc.
It’s good to have a heuristic feel for the …
I used to be very gung-ho about machine learning approaches to trading but I’m less so now. You have to understand that that there is a spectrum of alpha sources, from very specific structured arbitrage opportunities -> to stat arb -> to just voodoo nonsense.
As history goes on, hedge funds and other large players are absorbing the alpha from left to right. Having squeezed the pure arbs (ADR vs underlying, ETF vs components, mergers, currency triangles, etc) they then became hungry again and moved to stat arb (momentum, correlated pairs, regression analysis, news sentiment, etc). But now even the big stat arb strategies are running dry so people go further, chasing mirages (nonlinear regression, causality inference in large data sets, etc).
In modeling the market, it’s best to start with as much structure as possible before moving on to more amorphous statistical strategies. If you have to use statistical machine learning, encode as much trading domain knowledge as possible with specific distance/neighborhood metrics, linearity, variable importance weightings, hierarchy, low-dimensional factors, etc.
It’s good to have a heuristic feel for the danger/flexibility/noise sensitivity (synonyms) of each statistical learning tool. I roughly have this spectrum in my head:
Very specific, structured, safe
Optimize 1 parameter, require crossvalidation↓Optimize 2 parameters, require crossvalidation↓Optimize parameters with too little data, require regularization↓Extrapolation↓Nonlinear (SVM, tree bagging, etc)↓Higher-order variable dependencies↓Variable selection↓Structure learning
Very general, dangerous in noise, voodoo
This diagram is worth expanding. If anyone has any suggestions, please leave them.
Trending Now
1
Analytics
69 shares2,575 views
69 shares2,575 views
The Importance Of Creating A Single View of The Customer With Data
Clark Richey - March 11, 20192
Artificial Intelligence
33 shares1,789 views
33 shares1,789 views
How Artificial Intelligence Puts James Bond Tracking Tools In Your Home
Diana Hope - March 11, 20193
Big Data
55 shares3,044 views
55 shares3,044 views
How A Shortage of Data Scientists in The US is Holding Back Big Data
Andrej Kovacevic - March 8, 20194
Big Data
34 shares2,463 views
34 shares2,463 views
Machine Learning Makes VPNs Excellent Options For Cybersecurity
Diana Hope - March 8, 20195
Big Data
90 shares2,919 views
90 shares2,919 views
How Your Business Can Use Online Surveys to Gather Insights
Larry Alton - March 8, 20196
Big Data
43 shares2,573 views
43 shares2,573 views
Which Apps Use The Most Data, And What To Do About It
Sean Mallon - March 8, 20197
IT
1214 views
1214 views
You must log in to post a comment.