Portfolio Construction: From Theory to Practice

Module: Fintech AI | Difficulty: Advanced

Rebalancing

Transaction Cost

Turnover

Rebalancing Frequency

| Frequency | TC | Tracking Error | |-----------|-----|----------------| | Daily | High | Low | | Weekly | Medium | Low-Medium | | Monthly | Low | Medium | | Quarterly | Very Low | High |

import numpy as np

class PortfolioRebalancer:
    def __init__(self, target_weights, turnover_limit=0.1):
        self.target = target_weights; self.limit = turnover_limit
    def rebalance(self, current_weights, costs):
        desired = self.target
        turnover = np.abs(desired - current_weights).sum() / 2
        if turnover > self.limit:
            scale = self.limit / turnover
            desired = current_weights + scale * (desired - current_weights)
        tc = np.abs(desired - current_weights).dot(costs)
        return desired, tc

Research Insight: Practical portfolio construction differs from theory because of transaction costs, taxes, and constraints. The key insight is that frequent rebalancing improves tracking but increases costs. The optimal rebalancing frequency depends on the trade-off between these effects.