Adaptive Tracking-Error Control for Chinese Index-Enhancement Portfolios

Summary

This report describes a Chinese equity index-enhancement strategy built from a composite stock-selection signal and portfolio constraints. It combines factors spanning company size, valuation, growth, profitability, technical behavior, liquidity, and volatility. Symmetric orthogonalization is used to reduce overlap among factors, while a factor whose rolling estimated weight conflicts with its investment rationale is excluded from return forecasts.

The static portfolio optimizer seeks higher expected returns while matching the benchmark’s industry and size exposures and limiting each stock’s weight deviation from its benchmark weight. The adaptive version uses the recent relationship between those deviation limits and realized tracking error to set the next period’s limits, aiming to keep tracking error closer to a target as market conditions change. The report gives historical factor and portfolio performance figures for the CSI 300, CSI 500, and CSI 1000, including excess returns, drawdowns, information ratios, and adaptive-model tracking errors. These are reported results, not independent validation; the stated risks include broad market shocks and changes in factor effectiveness.

Key ideas

• The composite signal combines selected factors across several company and market characteristics.
• Symmetric orthogonalization is used to reduce multicollinearity among the selected factors.
• Factors are set aside when their estimated weight direction conflicts with their investment rationale.
• The optimizer controls industry and size exposures and limits stock weights relative to the benchmark.
• Recent realized tracking error informs the next period’s stock-weight deviation limits.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.