Regularized Fund Ratings and FOF Portfolio Construction

Summary

This report describes a regularized fund-evaluation model used to rank Chinese equity funds and form fund-of-funds portfolios. It compares funds selected for overall ability, stock-selection ability, and market-timing ability, then reviews their holdings and sector tilts. The reported profiles differ: overall-rated funds are relatively diversified, while stock-selection leaders show greater concentration and substantial exposure to the electric-vehicle supply chain. A high-score group also differs from a low-score group in its sector and stock allocations.

The report presents historical evidence, including portfolio comparisons and a decile backtest of a long-short portfolio built from composite scores. It reports positive average quarterly returns and a win rate above half, but provides no basis here to infer future performance. It also discusses estimation risk: coefficients fitted over roughly 60 trading days may fluctuate, and a local solution based on historical holdings reportedly estimates style exposures more accurately than a global solution, except for momentum. All findings are historical and remain sensitive to market and factor-return changes.

Key ideas

• The model ranks funds by composite performance, stock-selection ability, and timing ability.
• The report compares portfolio holdings, sector preferences, and differences between high- and low-scoring groups.
• A decile backtest is described as showing ordered performance and positive historical results for a score-based long-short portfolio.
• Short estimation windows can make fitted coefficients unstable.
• The report says local solutions improve style-exposure estimates for most factors, with momentum as an exception.

Tags

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