Machine Learning for Algorithmic Trading

Theexplosivegrowthofdigitaldatahasboostedthedemandforexpertiseintradingstrategiesthatusemachinelearning(ML).Thisrevisedandexpandedsecondeditionenablesyoutobuildandevaluatesophisticatedsupervised,unsupervised,andreinforcementlearningmodels.Thisbookintroducesend-to-endmachinelearningforthetradingworkflow,fromtheideaandfeatureengineeringtomodeloptimization,strategydesign,andbacktesting.Itillustratesthisbyusingexamplesrangingfromlinearmodelsandtree-basedensemblestodeep-learningtechniquesfromcuttingedgeresearch.Thiseditionshowshowtoworkwithmarket,fundamental,andalternativedata,suchastickdata,minuteanddailybars,SECfilings,earningscalltranscripts,financialnews,orsatelliteimagestogeneratetradeablesignals.ItillustrateshowtoengineerfinancialfeaturesoralphafactorsthatenableanMLmodeltopredictreturnsfrompricedataforUSandinternationalstocksandETFs.ItalsoshowshowtoassessthesignalcontentofnewfeaturesusingAlphalensandSHAPvaluesandincludesanewappendixwithoveronehundredalphafactorexamples.Bytheend,youwillbeproficientintranslatingMLmodelpredictionsintoatradingstrategythatoperatesatdailyorintradayhorizons,andinevaluatingitsperformance.