Exponential Smoothing Signals with Separate Long and Short Positions

Summary

This document describes an Expert Advisor that maintains independent buy and sell positions, each closed by a fixed take-profit or stop-loss. Its entry signal comes from a two-parameter exponential smoothing model: one state estimates the price level and another estimates its trend slope. The next value is forecast by adding the slope to the level. Linear regression ratios are suggested as possible initial values for those states.

The author proposes optimizing the long and short settings separately, using fixed lots while searching across history length, smoothing factors, and stop-loss values. The document reports that the system showed notable trading behavior on EURUSD H4 and D1 charts during strong trends, but provides no quantified performance results or detailed test methodology. It recommends adding a higher-timeframe indicator filter for entries. The described behavior and parameter ranges are specific to this implementation; the document does not establish robustness across markets or periods.

Key ideas

• The signal uses exponential smoothing to estimate price level and trend slope.
• The forecast for the next observation is the estimated level plus the estimated slope.
• Buy and sell positions are managed independently with fixed take-profit and stop-loss exits.
• The suggested optimization process tunes long and short settings separately using fixed lots.
• The author recommends a higher-timeframe indicator filter and does not provide quantified performance evidence.

Tags

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