# Estimating Forward-Looking Stock Correlations from Risk Factors

^{*}

## Abstract

**:**

## 1. Introduction

- (i)
- It is symmetric;
- (ii)
- All its elements lie within the $[-1,1]$ interval;
- (iii)
- It has unit diagonal;
- (iv)
- It is positive semi-definite (psd).

- (v)
- The matrix is ex ante free of arbitrage;
- (vi)
- The values have a realistic structure.

## 2. Discussion on Existing Models

- Equi-Correlations

- Local Equi-Correlations

- Adjusted Ex Post

- Skewness Approach

## 3. Solutions from Factor Structures

**Theorem**

**1.**

**Proof.**

**Lemma**

**1.**

**Proof.**

**Definition**

**1.**

**Definition**

**2.**

#### 3.1. Quantitative Approach: Computing Nearest Implied

#### 3.1.1. Formulating the Problem

**Lemma**

**2.**

**Proof.**

**Lemma**

**3.**

**Proof.**

**Lemma**

**4.**

**Proof.**

#### 3.1.2. Numerical Method

#### 3.2. Economic Approach

## 4. Empirical Experiment

^{®}Core™ i5-8250U CPU with 1.60 GHz, using the statistical programming software R.

_{avg}. The factor solely relying on implied data resulted by far in the highest fn

_{avg}. However, similarity to the target matrix plays an subordinate role here. Similar to that of fn

_{avg}, a reduction pattern can also be seen for $\tilde{\alpha}$ among hybrid models. In Section 3.2, $\tilde{\alpha}$ was defined as the weight of the boundary and $(1-\tilde{\alpha})$ as the weight of ${X}_{\mathbb{P}}$ inside the risk-neutral factor correlations ${X}_{\mathbb{Q}}$. This means that the larger the number of risk factors, the less modification was required to match the observed implied market variance in this empirical test. Thus, a larger k is likely to explain the hidden (true) implied correlation matrix better. This time, comparing the purely forward looking measure to the hybrid estimations shows that when implied data were used, less modification was required. Generally, as $\tilde{\alpha}$${}_{\mathrm{avg}}$ was close to zero across all five economic models, it can be concluded that correlation risk premia enter modestly in the factor-structured implied correlation matrix framework.

## 5. Concluding Remarks

## Author Contributions

## Funding

## Data Availability Statement

## Conflicts of Interest

## Appendix A. Non-Gaussian Copula: Example of Variance-Gamma

## References

- Skintzi, V.D.; Refenes, A.P.N. Implied correlation index: A new measure of diversification. J. Futur. Mark.
**2005**, 25, 171–197. [Google Scholar] [CrossRef] - Linders, D.; Schoutens, W. Basket option pricing and implied correlation in a one-factor Lévy model. In Innovations in Derivatives Markets; Springer: Cham, Switzerland, 2016; pp. 335–367. [Google Scholar]
- Milevsky, M.A.; Posner, S.E. A Closed-Form Approximation for Valuing Basket Options. J. Deriv.
**1998**, 5, 54–61. [Google Scholar] [CrossRef] - Harris, R.D.F.; Li, X.; Qiao, F. Option-implied betas and the cross section of stock returns. J. Futur. Mark.
**2019**, 39, 94–108. [Google Scholar] [CrossRef][Green Version] - Schadner, W. Ex-Ante Risk Factors and Required Structures of the Implied Correlation Matrix. Financ. Res. Lett.
**2021**, 41, 101855. [Google Scholar] [CrossRef] - Driessen, J.; Maenhout, P.J.; Vilkov, G. The Price of Correlation Risk: Evidence from Equity Options. J. Financ.
**2009**, 64, 1377–1406. [Google Scholar] [CrossRef][Green Version] - Fink, H.; Geppert, S. Implied correlation indices and volatility forecasting. Appl. Econ. Lett.
**2017**, 24, 584–588. [Google Scholar] [CrossRef] - Markopoulou, C.; Skintzi, V.; Refenes, A. On the predictability of model-free implied correlation. Int. J. Forecast.
**2016**, 32, 527–547. [Google Scholar] [CrossRef] - Buss, A.; Vilkov, G. Measuring Equity Risk with Option-implied Correlations. Rev. Financ. Stud.
**2012**, 25, 3113–3140. [Google Scholar] [CrossRef] - Kempf, A.; Korn, O.; Saßning, S. Portfolio Optimization Using Forward-Looking Information. Rev. Financ.
**2015**, 19, 467–490. [Google Scholar] [CrossRef][Green Version] - Hardle, W.K.; Silyakova, E. Implied basket correlation dynamics. Stat. Risk Model.
**2016**, 33, 1–20. [Google Scholar] [CrossRef][Green Version] - Chang, B.Y.; Christoffersen, P.; Jacobs, K.; Vainberg, G. Option-Implied Measures of Equity Risk. Rev. Financ.
**2011**, 16, 385–428. [Google Scholar] [CrossRef][Green Version] - Echaust, K.; Małgorzata, J. Implied correlation index: An application to economic sectors of commodity futures and stock markets. Eng. Econ.
**2020**, 31, 4–17. [Google Scholar] [CrossRef][Green Version] - Dhaene, J.; Dony, J.; Forys, M.B.; Linders, D.; Schoutens, W. FIX: The fear index—Measuring market fear. In Topics in Numerical Methods for Finance; Springer: Berlin/Heidelberg, Germany, 2012; pp. 37–55. [Google Scholar]
- Just, M.; Echaust, K. Stock market returns, volatility, correlation and liquidity during the COVID-19 crisis: Evidence from the Markov switching approach. Financ. Res. Lett.
**2020**, 37, 101775. [Google Scholar] [CrossRef] - González, M.T.; Novales, A. Are volatility indices in international stock markets forward looking? Rev. R. Acad. Cien. Ser. A Mat.
**2009**, 103, 339–352. [Google Scholar] [CrossRef][Green Version] - Kim, J.; Park, Y.J. Predictability of OTC option volatility for future stock volatility. Sustainability
**2020**, 12, 5200. [Google Scholar] [CrossRef] - Hollstein, F.; Prokopczuk, M.; Tharann, B.; Wese Simen, C. Predicting the equity market with option-implied variables. Eur. J. Financ.
**2019**, 25, 937–965. [Google Scholar] [CrossRef] - Buss, A.; Schoenleber, L.; Vilkov, G. Expected Correlation and Future Market Returns. SSRN Electron. J.
**2019**. [Google Scholar] [CrossRef] - Cecchetti, S.; Sigalotti, L. Forward-Looking Robust Portfolio Selection. Bank Italy Temi Discuss.
**2013**, 913. [Google Scholar] [CrossRef] - Dew-Becker, I. Real-Time Forward-Looking Skewness over the Business Cycle. Northwestern University Working Paper. 2021. Available online: http://www.dew-becker.org/ (accessed on 12 April 2022).
- Chamizo, Á.; Fonollosa, A.; Novales, A. Forward-looking asset correlations in the estimation of economic capital. J. Int. Financ. Mark. Institutions Money
**2019**, 61, 264–288. [Google Scholar] [CrossRef][Green Version] - Athanasakou, V.; Hussainey, K. The perceived credibility of forward-looking performance disclosures. Account. Bus. Res.
**2014**, 44, 227–259. [Google Scholar] [CrossRef][Green Version] - Conrad, J.; Dittmar, R.F.; Ghysels, E. Ex Ante Skewness and Expected Stock Returns. J. Financ.
**2013**, 68, 85–124. [Google Scholar] [CrossRef][Green Version] - Higham, N.J. Computing the nearest correlation matrix—a problem from finance. IMA J. Numer. Anal.
**2002**, 22, 329–343. [Google Scholar] [CrossRef][Green Version] - Sharpe, W.F. Capital Asset Prices: A Theory of Market Equilibrium under Conditions of Risk. J. Financ.
**1964**, 19, 425–442. [Google Scholar] - Fama, E.F.; French, K.R. Common risk factors in the returns on stocks and bonds. J. Financ. Econ.
**1993**, 33, 3–56. [Google Scholar] [CrossRef] - Fama, E.F.; French, K.R. A five-factor asset pricing model. J. Finac. Econ.
**2015**, 116, 1–22. [Google Scholar] [CrossRef][Green Version] - Numpacharoen, K.; Numpacharoen, N. Estimating Realistic Implied Correlation Matrix from Option Prices. J. Math. Financ.
**2013**, 3, 401–406. [Google Scholar] [CrossRef][Green Version] - Harvey, C.R.; Liu, Y.; Zhu, H. …and the Cross-Section of Expected Returns. Rev. Financ. Stud.
**2015**, 29, 5–68. [Google Scholar] [CrossRef][Green Version] - Jensen, T.I.; Kelly, B.T.; Pedersen, L.H. Is There a Replication Crisis in Finance? J. Financ.
**2022**. Forthcoming. [Google Scholar] - Glasserman, P.; Suchintabandid, S. Correlation expansions for CDO pricing. J. Bank. Financ.
**2007**, 31, 1375–1398. [Google Scholar] [CrossRef] - Borsdorf, R.; Higham, N.J.; Raydan, M. Computing a nearest correlation matrix with factor structure. SIAM J. Matrix Anal. Appl.
**2010**, 31, 2603–2622. [Google Scholar] [CrossRef][Green Version] - Guo, G.; Jacquier, A.; Martini, C.; Neufcourt, L. Generalized arbitrage-free SVI volatility surfaces. SIAM J. Financ. Math.
**2016**, 7, 619–641. [Google Scholar] [CrossRef][Green Version] - Buss, A.; Schönleber, L.; Vilkov, G. Option-Implied Correlations, Factor Models, and Market Risk. INSEAD Working Paper No 2017/20/FIN. 2017. Available online: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2906484 (accessed on 12 April 2022).
- Buraschi, A.; Porchia, P.; Trojani, F. Correlation Risk and Optimal Portfolio Choice. J. Financ.
**2010**, 65, 393–420. [Google Scholar] [CrossRef][Green Version] - Buraschi, A.; Trojani, F.; Vedolin, A. When Uncertainty Blows in the Orchard: Comovement and Equilibrium Volatility Risk Premia. J. Financ.
**2014**, 69, 101–137. [Google Scholar] [CrossRef] - Driessen, J.; Maenhout, P.J.; Vilkov, G. Option-Implied Correlations and the Price of Correlation Risk. Advanced Risk & Portfolio Management, SSRN. 2013. Available online: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2166829 (accessed on 12 April 2022).
- Faria, G.; Kosowski, R.; Wang, T. The correlation risk premium: International evidence. J. Bank. Financ.
**2022**, 136, 106399. [Google Scholar] [CrossRef] - Numpacharoen, K. Weighted Average Correlation Matrices Method for Correlation Stress Testing and Sensitivity Analysis. J. Deriv.
**2013**, 21, 67–74. [Google Scholar] [CrossRef] - Arellano-Valle, R.B.; Azzalini, A. The centred parametrization for the multivariate skew-normal distribution. J. Multivar. Anal.
**2008**, 99, 1362–1382. [Google Scholar] [CrossRef] - Barzilai, J.; Borwein, J.M. Two-point step size gradient methods. IMA J. Numer. Anal.
**1988**, 8, 141–148. [Google Scholar] [CrossRef] - Birgin, E.G.; Martinez, J.M.; Raydan, M. Spectral projected gradient methods: Review and perspectives. J. Stat. Softw.
**2014**, 60, 1–21. [Google Scholar] [CrossRef] - Birgin, E.G.; Martinez, J.M.; Raydan, M. Algorithm 813: SPG—Software for convex-constrained optimization. ACM Trans. Math. Softw. (TOMS)
**2001**, 27, 340–349. [Google Scholar] [CrossRef] - Martinez, J.M.; Pilotta, E.A. Inexact-Restoration Algorithm for Constrained Optimization1. J. Optim. Theory Appl.
**2000**, 104, 135–163. [Google Scholar] [CrossRef] - Gomes-Ruggiero, M.A.; Martinez, J.M.; Santos, S.A. Spectral Projected Gradient Method with Inexact Restoration for Minimization with Nonconvex Constraints. SIAM J. Sci. Comput.
**2009**, 31, 1628–1652. [Google Scholar] [CrossRef][Green Version] - Miller, K.S. On the Inverse of the Sum of Matrices. Math. Mag.
**1981**, 54, 67–72. [Google Scholar] [CrossRef] - Boyle, P.; Feng, S.; Melkuev, D.; Yang, S.; Zhang, J. Short Positions in the First Principal Component Portfolio. N. Am. Actuar. J.
**2018**, 22, 223–251. [Google Scholar] [CrossRef] - OptionMetrics. IvyDB US File and Data Reference Manual, 3rd ed.; OptionMetrics: New York, NY, USA, 2016. [Google Scholar]
- Ulrich, M.; Walther, S. Option-implied information: What’s the vol surface got to do with it? Rev. Deriv. Res.
**2020**, 23, 323–355. [Google Scholar] [CrossRef] - Etling, C.; Miller Thomas, W.J. The relationship between index option moneyness and relative liquidity. J. Futur. Mark.
**2000**, 20, 971–987. [Google Scholar] [CrossRef] - Carr, P.; Wu, L. The Finite Moment Log Stable Process and Option Pricing. J. Financ.
**2003**, 58, 753–777. [Google Scholar] [CrossRef][Green Version] - Ghalanos, A.; Theussl, S. Rsolnp: General Non-Linear Optimization Using Augmented Lagrange Multiplier Method. CRAN. 2015. Available online: https://cran.r-project.org/web/packages/Rsolnp/index.html (accessed on 12 April 2022).
- Ye, Y. Interior Algorithms for Linear, Quadratic, and Linearly Constrained Non-Linear Programming. Ph.D. Thesis, Stanford University, Stanford, CA, USA, 1987. [Google Scholar]
- Bakshi, G.; Kapadia, N.; Madan, D. Stock Return Characteristics, Skew Laws, and the Differential Pricing of Individual Equity Options. Rev. Finac. Stud.
**2003**, 16, 101–143. [Google Scholar] [CrossRef] - Madan, D.B.; Seneta, E. The variance gamma (VG) model for share market returns. J. Bus.
**1990**, 63, 511–524. [Google Scholar] [CrossRef] - Madan, D.B.; Carr, P.P.; Chang, E.C. The variance gamma process and option pricing. Rev. Financ.
**1998**, 2, 79–105. [Google Scholar] [CrossRef][Green Version]

**Figure 1.**Visualization of the constraints for the two-asset/one-factor/one-market-constraint case. The inequality constraint $\Omega $ spans the gray box of mathematically feasible solutions, and the blue line defines the solutions which satisfy the market constraint. As can be seen, the market constraint actually consists of two convex curves. Hence, two orthogonal projections of X onto $g(X)=0$ exist, but only one (i.e., $\dot{X}$) has minimum distance to X.

**Table 1.**The evaluation of existing models with respect to the developed requirements for realistic implied correlation matrices. A check mark ✓ indicates that the model is compliant with the condition. If it does not guarantee to comply, then it is marked with a ✗. A realistic solution has to fulfill all constraints (i)–(vi).

Mathematical | Economical | |||||
---|---|---|---|---|---|---|

(i) | (ii) | (iii) | (iv) | (v) | (vi) | |

Symmetric | |C_{ij}| ≤ 1 | Unit Diag. | Pos. Semi-Def. | Arbitrage-Free | Hetero. Struct. | |

Model: | ||||||

Equi-Correlations | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ |

Local Equi-Correlations | ✓ | ✓ | ✓ | ✗ | ✓ | ✗ |

Adj. Ex Post of [9], $AEP1$ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ |

Adj. Ex Post of [29], $AEP2$ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ |

Skewness Approach | ✓ | ✗ | ✓ | ✓ | ✗ | ✓ |

**Table 2.**Summary statistics of computing nearest (quantitative approach, Panels A–B) and risk-factor structured (economic approach, Panel C) implied correlation matrices on monthly option data. As for the quantitative approach, it turns out that computations were carried out in a small amount of time, converging towards the optimal solution within few iterations. The nearest factor-structured matrix can thus be used either as a stand-alone estimate, or as a tool to repair positive semi-definiteness and invertibility. Using the economic approach, implied as well as common risk factors from models, such as CAPM or Fama–French, can be used to estimate the implied correlation matrix.

k | A | t_{avg} | t_{sd} | fn_{avg} | fn_{sd} | |v.tol|_{avg} | |v.tol|_{max} | iter_{avg} | iter_{sd} | Index | |
---|---|---|---|---|---|---|---|---|---|---|---|

Panel A: | hist. matrix | ||||||||||

SP100 | 1 | hist. | 0.051 | 0.024 | 159.9 | 172.8 | 5.1 × ${10}^{-11}$ | 1.5 × ${10}^{-8}$ | 3.037 | 1.491 | SP100 |

SP100 | 3 | hist. | 0.106 | 0.050 | 111.0 | 168.3 | 1.3 $\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-8}$ | 6.2$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-7}$ | 4.819 | 2.137 | SP100 |

SP100 | 5 | hist. | 0.140 | 0.077 | 102.7 | 166.9 | 1.9$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-8}$ | 9.9$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-7}$ | 4.990 | 1.977 | SP100 |

mean-reverting matrix | |||||||||||

SP100 | 1 | m.r. | 0.055 | 0.027 | 158.0 | 158.2 | 2.0$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-11}$ | 2.5$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-9}$ | 3.064 | 1.438 | SP100 |

SP100 | 3 | m.r. | 0.107 | 0.047 | 113.3 | 173.2 | 3.9$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-8}$ | 9.6$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-7}$ | 4.708 | 1.790 | SP100 |

SP100 | 5 | m.r. | 0.149 | 0.082 | 104.6 | 172.5 | 2.4$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-8}$ | 8.9$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-7}$ | 5.054 | 1.989 | SP100 |

Panel B: | |||||||||||

SP500 | 1 | hist. | 5.347 | 2.311 | 2847.9 | 4395.0 | 1.5$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-11}$ | 1.3$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-10}$ | 5.003 | 1.892 | SP500 |

Repaired ${C}_{\mathbb{Q}}^{AEP1}$ | 15 | ${\mathrm{m}.\mathrm{r}.}_{\mathbb{Q}}$ | 0.223 | 0.070 | 16.41 | 6.775 | 3.1$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-8}$ | 9.7$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-7}$ | 7.184 | 2.351 | SP100 |

SQP | 1 | hist. | 0.337 | 0.098 | 161.4 | 166.8 | 3.5$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-7}$ | 9.6$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-7}$ | 3.517 | 0.721 | SP100 |

Panel C: | $\tilde{\alpha}$${}_{\mathrm{avg}}$ | $\tilde{\alpha}$${}_{\mathrm{sd}}$ | |||||||||

CAPM | 1 | skew${}_{\mathbb{Q}}$ | 0.018 | 0.014 | 4925.6 | 4548.0 | 5.0$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-17}$ | 4.4$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-16}$ | 0.099 | 0.083 | SP500 |

CAPM | 1 | hist. | 0.017 | 0.018 | 5108.5 | 5858.9 | 3.0$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-17}$ | 4.7$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-16}$ | 0.138 | 0.133 | SP500 |

Fama-Fr.3 | 3 | hist. | 0.019 | 0.018 | 4376.6 | 4740.8 | 3.3$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-17}$ | 2.5$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-16}$ | 0.091 | 0.070 | SP500 |

FF3+Mom. | 4 | hist. | 0.019 | 0.019 | 4167.1 | 4381.3 | 3.3$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-17}$ | 3.1$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-16}$ | 0.082 | 0.060 | SP500 |

Fama-Fr.5 | 5 | hist. | 0.019 | 0.024 | 4125.3 | 4509.1 | 3.6$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-17}$ | 3.9$\phantom{\rule{3.33333pt}{0ex}}\times \phantom{\rule{3.33333pt}{0ex}}{10}^{-16}$ | 0.076 | 0.055 | SP500 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Schadner, W.; Traut, J.
Estimating Forward-Looking Stock Correlations from Risk Factors. *Mathematics* **2022**, *10*, 1649.
https://doi.org/10.3390/math10101649

**AMA Style**

Schadner W, Traut J.
Estimating Forward-Looking Stock Correlations from Risk Factors. *Mathematics*. 2022; 10(10):1649.
https://doi.org/10.3390/math10101649

**Chicago/Turabian Style**

Schadner, Wolfgang, and Joshua Traut.
2022. "Estimating Forward-Looking Stock Correlations from Risk Factors" *Mathematics* 10, no. 10: 1649.
https://doi.org/10.3390/math10101649