Professor Doojin Ryu, Director of the Global Finance Research Center, worked with Dr. Geul Lee, a BK21 Research Professor, and Professor Li Yang of the University of New South Wales. They propose a new pre-processing method for options data that refines the dataset and increases the precision of return forecasts.

This study reveals that day‑to‑day changes in quote availability at extreme strike prices weaken predictive strength. To solve this problem, the team introduces Domain Stabilization, or DStab, an easy‑to‑apply rule that markedly enhances implied‑moment‑based forecasts.

Option quotes span a wide range of strike prices. However, quotes for deep out-of-the-money contracts, whose strikes lie far from the current underlying price and offer little chance of profit, are often missing or so low that they lack credibility. Because the range of strikes with reliable quotes changes over time, implied-moment estimates accumulate extra noise. This study proposes a noise reduction rule that discards quotes, even seemingly reliable ones, when they would otherwise introduce noise and distort the estimates. The resulting filter sharpens the information about future underlying returns embedded in the implied moments.

Empirical tests confirm that DStab outperforms earlier pre-processing methods. By removing option quotes according to rigorous thresholds, the approach strengthens return predictions and offers a practical answer to the persistent lack of reliable observations in derivatives markets.

※ Title: Domain stabilization for model-free option implied moment estimation

※ Journal: Journal of Financial Econometrics

※ Publisher: Oxford University Press

※ Author(s): First Author – Lee, G. (Geul Lee); Corresponding Author – Ryu, D. (Doojin Ryu); Co-Author – Yang, L.

※ DOI: https://doi.org/10.1093/jjfinec/nbae037