In this project, I built a realistic backtester that uses the Barra data. The backtester performs portfolio optimization that includes transaction costs, and was implemented with computational efficiency in mind, to allow for a reasonably fast backtest. I also used performance attribution to identify the major drivers of my portfolio's profit-and-loss (PnL).
willhughes11/Backtesting
Udacity AI for Trading - AI Algorithms for Trading: Backtesting
Jupyter Notebook