Volume-Based Linear Regression Signals Combined with MACD

Summary

This strategy estimates a price level from a linear regression of recent closing prices against trading volume, then combines that estimate with MACD conditions. A long signal requires the predicted level to fall between the bar’s open and close, to be rising, and for MACD measures to be above their signal lines. A short signal requires falling MACD conditions and the close below a risk reference based on the predicted level, optionally smoothed with a volume-weighted moving average.

The document proposes regression smoothing, stop losses, sentiment inputs, machine learning, and confirmation across timeframes as possible extensions. It provides no performance evidence for the claimed predictive approach. It also acknowledges that historical regression can miss sudden news and that results may depend on lookback choices. The published settings describe a one-week, one-minute BTC/USDT futures test, too narrow to support broad conclusions.

Key ideas

• The method fits recent prices against volume to produce a predicted price reference.
• Long entries combine the predicted level’s position and upward movement with bullish MACD conditions.
• Short entries use declining MACD conditions and a close below the selected predicted-price risk reference.
• The document notes sensitivity to regression settings and the possibility of missing sudden news-driven moves.
• The published backtest covers one week of one-minute BTC/USDT futures data and reports no performance results.

Tags

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