Using FMZ Time-Series Formulas to Explore and Plot Trading Signals

Summary

This guide describes an analysis tool for calculating and plotting expressions over market time series, with syntax influenced by the public Alpha101 formula collection. It lists arithmetic and conditional operators alongside functions for lagging observations, moving averages, correlations, covariance, scaling, changes, weighted decay, rolling extrema, ranks, sums, products, and standard deviations. These primitives let a researcher compose and visually inspect candidate signals.

Available inputs include returns, OHLC prices, volume, and instrument selection; multiple expressions can be plotted together. Cross-sectional ranking depends on specifying a pool of instruments, while the guide says industry neutralization is unsupported. Several fields, including VWAP, market capitalization, and industry classification, are marked unimplemented, with VWAP temporarily treated as close. The tool is presented for rapid idea checks, not as evidence that a signal is profitable; the guide supplies no validated strategy, backtest results, or execution assumptions.

Key ideas

• The tool evaluates formulas over selected market time series and displays their results.
• Its function set includes rolling statistics, time-series ranks, correlations, scaling, and moving averages.
• Cross-sectional ranks require a multi-instrument candidate pool.
• Some inputs and functions are unavailable or approximated, including industry neutralization and VWAP.
• Visualizing a formula helps inspect an idea but does not establish its trading performance.

Tags

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