Using EMD Signal-to-Noise Ratios to Improve Equity Reversal Factors

Summary

This research summary treats intraday stock prices as time series rather than isolated observations. It applies empirical mode decomposition (EMD) to separate trend-like price signals from noise, then uses their relative strength to form a signal-to-noise ratio (SNR) factor. Stocks with stronger trends and less noise are reported to perform better in the following month.

The report then combines the SNR measure with a reversal factor, arguing that filtering noisy price movements makes reversal signals more informative. The supplied summary reports stronger information coefficients and improved long-only returns for the refined factor compared with the original reversal measure, alongside performance statistics for the SNR portfolio. These findings are reported figures, but the underlying paper, sample construction, transaction costs, and robustness checks are not included here; the summary alone is insufficient to assess out-of-sample reliability.

Key ideas

• The study uses EMD to separate intraday stock-price movements into signal and noise components.
• It constructs an SNR factor to identify stocks with stronger, less noisy trends.
• The summary reports that high-SNR stocks perform better in the following month.
• Combining SNR with reversal is reported to strengthen the reversal factor’s measured performance.
• The supplied text omits the full report and details needed to assess robustness or trading costs.

Tags

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