Batch Factor Evaluation with Adaptive Direction and Turnover Costs
Summary
The document describes a tool for evaluating many equity factors in batches and saving overall and yearly long-side results. Users configure the factor source and names, stock universe, benchmark, optional smoothing, and output settings. For high-frequency factors, it suggests trying a 20-day smoothing window, which the author says can reduce turnover and make costs easier to assess; no supporting performance data is provided.
The tool selects factor direction by comparing grouped cumulative returns and can estimate annualized net returns by subtracting costs inferred from annualized turnover. The author cautions that this cost calculation is a simple estimate intended as a general reference. The described SQL query does not handle joins across tables without modification, and the document gives no detailed methodology for grouping, cost assumptions, or validation of results.
Key ideas
- Batch evaluation can speed comparisons across a large set of factors.
- Factor direction is chosen by comparing cumulative returns across groups.
- Smoothing a high-frequency factor over 20 days may reduce turnover and make its performance easier to evaluate after costs.
- Annualized turnover is used to estimate trading costs and net annualized returns.
- The cost model is approximate, and the SQL query requires changes for cross-table data.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.