Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Dynamic Conditional SKEPTIC

Gabriele Di Luzio, Giacomo Morelli

TL;DR

The paper addresses the limitation of normality in multivariate volatility models by introducing the Dynamic Conditional SKEPTIC (DCS), a semiparametric framework that uses nonparametric rank-based measures (Spearman's rho and Kendall's tau) within a Gaussian copula to model time-varying correlations. It combines a nonparanormal margin with a dynamic correlation process, estimated via a two-step composite likelihood to handle high dimensionality, and establishes stationarity, mixing properties, and an $O_P(\sqrt{\log(Tp)/T})$ convergence rate for the correlation estimator. Through simulations and an empirical study on S&P100/SP500 data (2013–2025), DCS shows improved diagnostic performance, robust parameter recovery under contamination, and practical portfolio benefits, including lower turnover and competitive Sharpe ratios. The results suggest that rank-based, semiparametric dependence modeling offers a robust and scalable alternative to classic DCC approaches, with meaningful implications for portfolio construction and risk management.

Abstract

We introduce the Dynamic Conditional SKEPTIC (DCS), a semiparametric approach for efficiently and robustly estimating time-varying correlations in multivariate models. We exploit nonparametric rank-based statistics, namely Spearman's rho and Kendall's tau, to estimate the unknown correlation matrix and discuss the stationarity, beta- and rho- mixing conditions of the model. We illustrate the methodology by estimating the time-varying conditional correlation matrix of the stocks included in the S&P100 and S&P500 during the period from 02/01/2013 to 23/01/2025. The results show that DCS improves diagnostic checks compared to the classical Dynamic Conditional Correlation (DCC) models, providing uncorrelated and normally distributed residuals. A risk management application shows that global minimum variance portfolios estimated using the DCS model exhibit lower turnover than those based on the DCC and DCC-NL models, while also achieving higher Sharpe ratios for portfolios constructed from S&P 100 constituents.

Dynamic Conditional SKEPTIC

TL;DR

The paper addresses the limitation of normality in multivariate volatility models by introducing the Dynamic Conditional SKEPTIC (DCS), a semiparametric framework that uses nonparametric rank-based measures (Spearman's rho and Kendall's tau) within a Gaussian copula to model time-varying correlations. It combines a nonparanormal margin with a dynamic correlation process, estimated via a two-step composite likelihood to handle high dimensionality, and establishes stationarity, mixing properties, and an convergence rate for the correlation estimator. Through simulations and an empirical study on S&P100/SP500 data (2013–2025), DCS shows improved diagnostic performance, robust parameter recovery under contamination, and practical portfolio benefits, including lower turnover and competitive Sharpe ratios. The results suggest that rank-based, semiparametric dependence modeling offers a robust and scalable alternative to classic DCC approaches, with meaningful implications for portfolio construction and risk management.

Abstract

We introduce the Dynamic Conditional SKEPTIC (DCS), a semiparametric approach for efficiently and robustly estimating time-varying correlations in multivariate models. We exploit nonparametric rank-based statistics, namely Spearman's rho and Kendall's tau, to estimate the unknown correlation matrix and discuss the stationarity, beta- and rho- mixing conditions of the model. We illustrate the methodology by estimating the time-varying conditional correlation matrix of the stocks included in the S&P100 and S&P500 during the period from 02/01/2013 to 23/01/2025. The results show that DCS improves diagnostic checks compared to the classical Dynamic Conditional Correlation (DCC) models, providing uncorrelated and normally distributed residuals. A risk management application shows that global minimum variance portfolios estimated using the DCS model exhibit lower turnover than those based on the DCC and DCC-NL models, while also achieving higher Sharpe ratios for portfolios constructed from S&P 100 constituents.

Paper Structure

This paper contains 24 sections, 4 theorems, 62 equations, 1 figure, 10 tables.

Table of Contents

  1. Introduction
  2. Background
  3. Notation
  4. Dynamic Conditional Corelation
  5. The Dynamic Conditional SKEPTIC Model
  6. Nonparanormal
  7. Dynamic Conditional SKEPTIC
  8. Theoretical Properties
  9. Mixing Conditions of Semiparametric Gaussian Process
  10. Estimation
  11. Asymptotic Properties of Composite Likelihood Estimator
  12. Simulation Studies
  13. Empirical Analysis
  14. Model Fit and Diagnostic Checks
  15. Portfolio Construction
  16. ...and 9 more sections

Key Result

Lemma 3.1

Assume that $\boldsymbol{X}_t \sim \operatorname{NPN}(0, H_t, f)$, where $H_t = D_t R_t D_t$. Then the elements of the correlation matrix $R_t$ satisfy: where $\rho^s_{ij,t}$ and $\tau_{ij,t}$ denote Spearman's rho and Kendall's tau between the variables $X_{it}$ and $X_{jt}$, respectively.

Figures (1)

  • Figure 1: Backtesting results for SP100 and SP500 across the DCC, DCC-NL, DCS (rho), DCS (tau) and EW specifications. Return observations are plotted against their corresponding VaR and ES estimates, with violations highlighted in green.

Theorems & Definitions (7)

  • Lemma 3.1
  • Definition 3.2
  • Proposition 4.1
  • Definition 4.2
  • Proposition 4.3
  • Definition 4.4
  • Proposition 4.5