/StockML

Research using machine learning to pick stocks

Primary LanguagePython

StockML

Research using machine learning to pick stocks

Initial Research

  • VIX - Fear Index
  • IBM
  • BAC - Bank of America
  • AGNC - REIT
  • GSPC - S&P 500
  • VTI - Vanguard Total Stock Market Index

Instructions

  1. Download quote files from yahoo (1/2/1990 - Present) Note that training set will begin on 1/2/1992 with a 20 yr window
  2. Rename file to ticker.csv
  3. python parse.py ticker.csv

Notes

  • TradeCost = 6.95 / trade (+ $0.75/option contract in the case of options) Per ShareBuilder
  • Profit = Range * NumShares - 2 * TradeCost
  • Minimum Number of Shares to Break Event: NumShares = (2 * TradeCost)/Range
  • Minimum investment to realize gain: MinInvestment = (2 * TradeCost)/Range * BuyPrice