Eigen-Coordinates for Identifying Q-Gaussian Return Distributions

Summary

The article introduces the eigen-coordinates method as a way to test how well candidate functions describe observed data. It constructs a basis from a function’s differential structure, expresses the data as a linear expansion in that basis, and estimates expansion coefficients with least squares. The proposed diagnostic is whether observations form the expected linear structure in the candidate function’s eigen-coordinates; a different generating function should produce a different structure.

The financial example concerns heavy-tailed return distributions and Q-Gaussians. The article applies its method to an S&P 500 daily-return sample and reports an estimated q near 1.55, compared with a cited book value near 1.4, concluding that the Q-Gaussian gives a reasonable description. It also notes that return distributions may approach normality at longer horizons and that a distribution can be numerically close to a Q-Gaussian without being analytically identical. The example is descriptive and does not establish that the fitted distribution forecasts returns or improves trading decisions.

Key ideas

• The method builds a function-specific coordinate basis from the differential structure of a candidate model.
• Least squares estimates the coefficients of the data’s expansion in that basis.
• The authors propose linearity in the candidate basis as a tool for distinguishing competing function forms.
• The S&P 500 daily-return example finds a Q-Gaussian fit with an estimated q near 1.55.
• A close numerical fit does not prove analytical identity or predictive trading value.

Tags

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