Estimating Price-Movement Probabilities and Trend Quality for Trading

Summary

This article proposes a price-movement model that estimates upward and downward step probabilities from historical price changes. It derives a normalized price speed from an average price change over a chosen interval, then uses that quantity to estimate directional probabilities and project a future price distribution. For forecasting normalized speed, it favors Fourier extrapolation of an oscillatory series over a simple conditional-average forecast, which the author argues would tend toward zero and produce few useful signals.

The article also defines trend quality as the expected price shift relative to its uncertainty, using the measure to reason about averaging periods and the horizon for trend-following trades. It argues that even a high-quality trend may end unexpectedly, so it favors taking profits on smaller favorable moves. The framework is theoretical and relies on approximate probabilities derived from observed data; the text provides no independent performance validation. Its claims about market dynamics and optimal parameters should therefore be treated as hypotheses rather than established results.

Key ideas

• The model estimates directional price-step probabilities using historical price movement over an averaging interval.
• Normalized price speed links average price change to estimated upward and downward probabilities.
• The author proposes Fourier extrapolation to forecast the oscillatory normalized-speed series.
• Trend quality compares expected directional movement with the uncertainty of that movement.
• The article cautions that a strong trend can reverse unexpectedly and offers no independent strategy validation.

Tags

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