Portfolio rebalancing restores intended asset weights after market movements cause allocation drift. It limits concentration risk and keeps decisions aligned with the original strategy. section 7.6.3
Choose a trigger and account for implementation costs:
- Periodic rebalancing acts at predefined intervals, such as quarterly. section 7.6.3
- Threshold-based rebalancing acts only when an asset deviates beyond a limit, such as ±5%. section 7.6.3
- A hybrid approach combines scheduled reviews with deviation thresholds, while limiting unnecessary transaction costs and possible tax consequences. section 7.6.3
A Python rebalancing workflow
Represent target weights and current dollar values as dictionaries. Compute total value and current allocations, subtract target weights to find deviations, filter deviations beyond the threshold, and calculate target value minus current value. A positive adjustment means Buy; a negative adjustment means Sell. A pandas DataFrame can then show current value, target value, adjustment, and action. The supplied example produces large sales of AAPL, WMT, GOOGL, and BRK-B and purchases of PFE and NVDA, illustrating why allocation rules should be checked for unintended de-diversification. section 7.6.3
target_allocation = {
"AAPL": 0.016804,
"WMT": 0.007427,
"GOOGL": 0.015956,
"KO": 0.142700,
"PFE": 0.193730,
"BRK-B": 0.009315,
"NVDA": 0.614068
}
Inspect the target allocation used by the calculation, including the unusually large NVDA weight that drives the example’s concentration risk. section 7.6.3
