Backtesting evaluates an investment strategy by simulating it over historical market data. For portfolio rebalancing, the simulation advances one day at a time, grows holdings using prices and dividends, generates trades, and records taxes, trading costs, and deviations from target weights. section 13.4 section 13.4.2
A practical backtester typically:
- Retrieves non-adjusted closing prices and dividend data, preserving dividends as cash rather than automatically reinvesting them. section 13.4.2
- Summarizes annualized return, volatility, turnover, spread cost, tax cost, rebalance frequency, and deviations from target weights. section 13.4.4
Comparing three rebalancing policies
A portfolio with six assets, $10,000 of starting capital, 0.2% target cash, and 0.03% spreads can be tested over 2013–2022. Weekly interval rebalancing produced 7.00% mean return, 39.8% turnover, and 0.21% mean maximum deviation; threshold rebalancing produced 7.04%, 31.9%, and 0.26%; optimization produced 7.04%, 17.6%, and 0.60%. The comparison exposes trade-offs among trading activity, taxes, and tracking precision. section 13.4.3
