Reconstructing an economic space from a market metric

TL;DR

This paper introduces a method to reconstruct an economic space from market data using a correlation-based metric, identifying a systematic subspace linked to market factors and relating network topology to market shocks.

Contribution

It proposes a novel approach to reconstruct economic space from market metrics and relates the subspace dimension to factor models, with empirical analysis on major stock indexes.

Findings

Identified a systematic subspace related to market factors.

Defined and empirically related continuous clustering to market shocks.

Demonstrated the method on DJIA and S&P500 data.

Abstract

Using a metric related to the returns correlation, a method is proposed to reconstruct an economic space from the market data. A reduced subspace, associated to the systematic structure of the market, is identified and its dimension related to the number of terms in factor models. Example were worked out involving sets of companies from the DJIA and S&P500 indexes. Having a metric defined in the space of companies, network topology coefficients may be used to extract further information from the data. A notion of "continuous clustering" is defined and empirically related to the occurrence of market shocks.