New Multi-factor Equity Index Algorithm

~ Stephen Harlin, MD

An Algorithmic System for Equity Index Investing that Combines Time-Series Momentum, DIX-GEX Scenarios, and Auction Market Dynamics

 

Multi-factor Model:

  • Trend following (TSMOM)
  • DIX-GEX scenarios
  • Auction market dynamics

Summary

TSMOM performs well on on aggregate indices (e.g. S&P 500). In an attempt to avoid prediction delay, the system employs the higher order ‘fold’ function to create an “instantaneous” moving average (iMA). iMA signals (crossovers and overnight gaps) are filtered by a moving linear regression calculation and then plotted on two charts: SPX 1-week and SPX daily.

At the same time, daily DIX-GEX data is followed for scenarios consistent with forced buying (re-hedging for negative gamma exposure) and anticipatory selling (DIX data below threshold and clustering at rally highs).

The final step in signal validation considers auction market process dynamics to ensure demand at LOW key reference levels (e.g. “buying tails”) and evidence for liquidation at HIGH key reference levels (e.g. “selling tails”).