/backtesting

Equity portfolio backtesting prototype, with an emphasis on computational efficiency

Primary LanguageJupyter Notebook

Backtesting

This projects builds a computationally effecient backtester for a market neutral US Equities portfolio using Barra data for the period 2006-2007.

Alpha factors:

  • USFASTD_1DREVRSL : Reversal

  • USFASTD_EARNYILD : Earnings Yield

  • USFASTD_VALUE : Value

  • USFASTD_SENTMT : Sentiment

drawing

Focus is on:

  • portfolio optimization accounting for transaction costs

drawing

  • performance attribution to identify the major drivers portfolio's PnL

drawing