Abstract
The Cornish–Fisher expansion is a simple way to determine quantiles of non-normal distributions. It is frequently used by practitioners and by academics in risk management, portfolio allocation, and asset liability management. It allows us to consider non-normality and, thus, moments higher than the second moment, using a formula in which terms in higher-order moments appear explicitly. This paper has two primary objectives. First, we resolve the classic confusion between the skewness and kurtosis coefficients of the formula and the actual skewness and kurtosis of the distribution when using the Cornish–Fisher expansion. Second, we use the response surface approach to estimate a function for these two values. This helps to overcome the difficulties associated with using the Cornish–Fisher expansion correctly to compute value at risk. In particular, it allows a direct computation of the quantiles. Our methodology has many practical applications in risk management and asset allocation.
Similar content being viewed by others
Notes
In addition, Aboura and Maillard (2016) use the Cornish–Fisher expansion to revisit the pricing of options, in a context of financial stress, when the underlying asset’s returns display skewness and excess kurtosis. They derive an exact formula allowing for heavy tails.
Other possibilities exist, but are not covered in this context.
In terms of gains rather than losses, the VaR at confidence level \(\alpha \) for a market rate of return X, with a distribution function \(F_X(x)\equiv P \left[ X \le x \right] \) and quantile at level \(\alpha \) denoted as \(q_\alpha (X)\), is
$$\begin{aligned} -{{ VaR}}_\alpha (X) = \sup \left\{ x:F_X(x) \le \alpha \right\} \equiv q_\alpha (X). \end{aligned}$$This approximation is based on the Taylor series developed, for example, by Stuart and Ord (2009).
At the third order, the approximation is: \(\forall \alpha \in (0,1), z_{{ CF},\alpha }=z_{\alpha }+\frac{1}{6}(z_{\alpha }^2-1)S \).
It is straightforward to show that in the presence of an underlying Gaussian distribution (\(S=0\) and \(K=3\)), Eq. (4) reduces to the Gaussian quantile. Thus, the Cornish–Fisher expansion can obviously be used when the distribution is normal.
Following Maillard (2012), \(\sigma _{{ CF}}=\dfrac{\sigma }{\sqrt{1+\dfrac{1}{96}K^2+\dfrac{25}{1296}S^4-\dfrac{1}{36}KS^2}}\).
However, exact distributions have advantages as well: they enable Monte Carlo simulations and, thus, allow for the direct computation of \({ VaR}\).
The 0.5% VaR of the Solvency II regulation requires a minimum of 17 years of data (17 years \(=\) 204 months).
Maillard (2012) computes the moments of the fourth-order Cornish–Fisher inverse expansion. The equations presented here correct a misprint and write S and K as functions of \(S_c\) and \(K_c\)
Surprisingly few studies systematically compare the performances of these optimization methods.
See also the chapter (by Jaschke and Jiang) of Hardle (2009) for a detailed discussion.
Standard RSM models usually include repeated powers and log transformation.
Although polynomials are popular in data analyses, linear and quadratic functions are severely limited in their range of curve shapes, whereas cubic and higher-order curves often produce undesirable characteristics, such as edge effects and waves (see Sauerbrei et al. 2007).
References
Aboura, S., & Maillard, D. (2016). Option pricing under skewness and kurtosis using a Cornish–Fisher expansion. Journal of Futures Markets, 36(12), 1194–1209.
Acerbi, C., & Tasche, D. (2002a). Expected shortfall: A natural coherent alternative to value at risk. Economic Notes, 31(2), 379–388.
Acerbi, C., & Tasche, D. (2002b). On the coherence of expected shortfall. Journal of Banking and Finance, 26(7), 1487–1503.
Adcroft, A., & Mason, R. B. (2007). The external environment’s effect on management and strategy: A complexity theory approach. Management Decision, 45(1), 10–28.
Ahmad, A., Wong, S., Teng, T., & Zuhairi, A. (2007). Optimization of coagulation–flocculation process for pulp and paper mill effluent by response surface methodological analysis. Journal of Hazardous Materials, 145(1), 162–168.
Amédée-Manesme, C.-O., Barthélémy, F., & Keenan, D. (2015). Cornish fisher expansion for commercial real estate value at risk. The Journal of Real Estate Finance and Economics, 50(4), 439–464.
Artzner, P., Delbaen, F., Eber, J.-M., & Heath, D. (1999). Coherent measures of risk. Mathematical Finance, 9(3), 203–228.
Bao, Y., Lee, T.-H., & Saltoglu, B. (2006). Evaluating predictive performance of value-at-risk models in emerging markets: A reality check. Journal of Forecasting, 25(2), 101–128.
Bezerra, M. A., Santelli, R. E., Oliveira, E. P., Villar, L. S., & Escaleira, L. A. (2008). Response surface methodology (RSM) as a tool for optimization in analytical chemistry. Talanta, 76(5), 965–977.
Box, G. E. & Draper, N. R. (1987). Introduction to response surface methodology. In: Response surfaces, mixtures, and ridge analyses, 2nd edn (pp. 1–16).
Box, G. E. & Wilson, K. B. (1992). On the experimental attainment of optimum conditions. In Breakthroughs in statistics (pp. 270–310). Springer.
Chernozhukov, V., Fernandez-Val, I., & Galichon, A. (2009). Improving point and interval estimators of monotone functions by rearrangement. Biometrika, 96(3), 559–575.
Chernozhukov, V., Fernandez-Val, I., & Galichon, A. (2010). Rearranging Edgeworth–Cornish–Fisher expansions. Economic Theory, 42(2), 419–435.
Christoffersen, P. F. (2012). Elements of financial risk management. London: Academic Press.
Cornish, E. A., & Fisher, R. A. (1938). Moments and cumulants in the specification of distributions. Revue de l’Institut International de Statistique/Review of the International Statistical Institute, 5(4), 307.
Del Castillo, E. (2007). Process optimization: A statistical approach (Vol. 105). Berlin: Springer.
Engle, R. F., & Manganelli, S. (2004). Caviar: Conditional autoregressive value at risk by regression quantiles. Journal of Business and Economic Statistics, 22(4), 367–381.
Fallon, W. (1996). Calculating value-at-risk (working paper no. 96-49).
Feuerverger, A., & Wong, A. C. (2000). Computation of value-at-risk for nonlinear portfolios. Journal of Risk, 3, 37–56.
Giunta, A., McFarland, J., Swiler, L., & Eldred, M. (2006). The promise and peril of uncertainty quantification using response surface approximations. Structures and Infrastructure Engineering, 2(3–4), 175–189.
Gunst, R. F. (1996). Response surface methodology: Process and product optimization using designed experiments. Routledge: Taylor and Francis Group.
Hardle, W. K. (2009). Applied quantitative finance. Berlin: Springer.
He, Z., Zhu, P.-F., & Park, S.-H. (2012). A robust desirability function method for multi-response surface optimization considering model uncertainty. European Journal of Operational Research, 221(1), 241–247.
Hill, W. J., & Hunter, W. G. (1966). A review of response surface methodology: A literature survey. Technometrics, 8(4), 571–590.
Ibrahim, H. M., & Elkhidir, E. E. (2011). Response surface method as an efficient tool for medium optimisation. Trends in Applied Sciences Research, 6(2), 121.
Isukapalli, S., Roy, A., & Georgopoulos, P. (2000). Efficient sensitivity/uncertainty analysis using the combined stochastic response surface method and automated differentiation: Application to environmental and biological systems. Risk Analysis, 20(5), 591–602.
Jacobson, S. H., & Schruben, L. W. (1989). Techniques for simulation response optimization. Operations Research Letters, 8(1), 1–9.
Jaschke, S. R. (2001). The Cornish–Fisher–Expansion in the context of delta–gamma–normal approximations [working paper].
Johnson, N. L. (1949). Bivariate distributions based on simple translation systems. Biometrika, 36(3/4), 297–304.
Jorion, P. (2007). Value at risk: The new benchmark for managing financial risk. New York, NY: McGraw-Hill.
Khataee, A. (2010). Optimization of UV-promoted peroxydisulphate oxidation of CI basic blue 3 using response surface methodology. Environmental Technology, 31(1), 73–86.
Khuri, A. I., & Cornell, J. A. (1996). Response surfaces: Designs and analyses (Vol. 152). London: CRC Press.
Khuri, A. I., & Mukhopadhyay, S. (2010). Response surface methodology. Wiley Interdisciplinary Reviews: Computational Statistics, 2(2), 128–149.
Lee, J., Ye, L., Landen, W. O., & Eitenmiller, R. R. (2000). Optimization of an extraction procedure for the quantification of vitamin E in tomato and broccoli using response surface methodology. Journal of Food Composition and Analysis, 13(1), 45–57.
Li, H., Zhou, S., Sun, Y., & Lv, J. (2010). Application of response surface methodology to the advanced treatment of biologically stabilized landfill leachate using fentons reagent. Waste Management, 30(11), 2122–2129.
Longin, F. M. (2000). From value at risk to stress testing: The extreme value approach. Journal of Banking and Finance, 24(7), 1097–1130.
MacKinnon, J. G. (2010). Critical values for cointegration tests (Technical report). Queen’s Economics Department working paper.
Maillard, D. (2012). A user’s guide to the Cornish–Fisher expansion. Available at SSRN 1997178.
Mina, J. & Ulmer, A. (1999). Delta–Gamma four ways (Vol. 1st quarter). MSCI.
Montgomery, D. C. (2001). Response surface methods and other approaches to process optimization. Design and analysis of experiments. New York: Wiley.
Muhamad, M. H., Abdullah, S. R. S., Mohamad, A. B., Rahman, R. A., & Kadhum, A. A. H. (2013). Application of response surface methodology (RSM) for optimisation of COD, NH\(_3\)–N and 2,4-DCP removal from recycled paper wastewater in a pilot-scale granular activated carbon sequencing batch biofilm reactor (GAC-SBBR). Journal of Environmental Management, 121, 179–190.
Myers, R. H. (1999). Response surface methodology-current status and future directions. Journal of Quality Technology, 31(1), 30.
Myers, R. H., Montgomery, D. C., & Anderson-Cook, C. M. (2016). Response surface methodology: Process and product optimization using designed experiments. London: Wiley.
Myers, R. H., Montgomery, D. C., Vining, G. G., Borror, C. M., & Kowalski, S. M. (2004). Response surface methodology: A retrospective and literature survey. Journal of Quality Technology, 36(1), 53.
Naguez, N., & Prigent, J.-L. (2017). Optimal portfolio positioning within generalized Johnson distributions. Quantitative Finance, 17, 1037–1055.
Nath, A., & Chattopadhyay, P. (2007). Optimization of oven toasting for improving crispness and other quality attributes of ready to eat potato-soy snack using response surface methodology. Journal of Food Engineering, 80(4), 1282–1292.
Neddermeijer, H. G., Van Oortmarssen, G. J., Piersma, N. & Dekker, R. (2000). A framework for response surface methodology for simulation optimization. In Proceedings of the 32nd conference on winter simulation (pp. 129–136).
Nwabueze, T. U. (2010). Review article: Basic steps in adapting response surface methodology as mathematical modelling for bioprocess optimisation in the food systems. International Journal of Food Science and Technology, 45(9), 1768–1776.
Pichler, S. & Selitsch, K. (1999). A comparison of analytical var methodologies for portfolios that include options [Working paper].
Prasad, K. N., Hassan, F. A., Yang, B., Kong, K. W., Ramanan, R. N., Azlan, A., et al. (2011). Response surface optimisation for the extraction of phenolic compounds and antioxidant capacities of underutilised Mangifera pajang Kosterm. peels. Food Chemistry, 128(4), 1121–1127.
Pritsker, M. (1997). Evaluating value at risk methodologies: Accuracy versus computational time. Journal of Financial Services Research, 12(2–3), 201–242.
Rockafellar, R. T., & Uryasev, S. (2002). Conditional value-at-risk for general loss distributions. Journal of Banking and Finance, 26(7), 1443–1471.
Royston, P., & Altman, D. G. (1994). Regression using fractional polynomials of continuous covariates: Parsimonious parametric modelling. Applied Statistics, 43, 429–467.
Royston, P., & Sauerbrei, W. (2008). Multivariable model-building: A pragmatic approach to regression anaylsis based on fractional polynomials for modelling continuous variables (Vol. 777). London: Wiley.
Salmasnia, A., Moeini, A., Mokhtari, H., & Mohebbi, C. (2013). A robust posterior preference decision-making approach to multiple response process design. International Journal of Applied Decision Sciences, 6(2), 186–207.
Sauerbrei, W., & Royston, P. (1999). Building multivariable prognostic and diagnostic models: Transformation of the predictors by using fractional polynomials. Journal of the Royal Statistical Society: Series A (Statistics in Society), 162(1), 71–94.
Sauerbrei, W., Royston, P., & Binder, H. (2007). Selection of important variables and determination of functional form for continuous predictors in multivariable model building. Statistics in Medicine, 26(30), 5512–5528.
Simonato, J.-G. (2011). The performance of Johnson distributions for computing value at risk and expected shortfall. The Journal of Derivatives, 19(1), 7–24.
Stuart, A., & Ord, K. (2009). Kendall’s advanced theory of statistics: Volume 1: Distribution theory. London: Wiley.
Weng, W.-L., Liu, Y.-C., & Lin, C.-W. (2001). Studies on the optimum models of the dairy product Kou Woan Lao using response surface methodology. Asian Australasian Journal of Animal Sciences, 14(10), 1470–1476.
Yeo, I.-K., & Johnson, R. A. (2000). A new family of power transformations to improve normality or symmetry. Biometrika, 87(4), 954–959.
Yu, P. L., Li, W. K., & Jin, S. (2010). On some models for value-at-risk. Econometric Reviews, 29(5–6), 622–641.
Zangari, P. (1996). A var methodology for portfolios that include options (Vol. 4th quarter). MSCI.
Acknowledgements
Charles-Olivier Amédée-Manesme would like to acknowledge the financial support of Chaire d’assurance et de services financiers L’Industrielle-Alliance, at FSA ULaval and of the Fonds de Recherche du Québec - Société et Culture (Grant FRQSC 2016-NP-190494).
Author information
Authors and Affiliations
Corresponding author
Appendices
A Appendix: Quantile functions
The quantile function (or inverse cumulative distribution function) of the probability distribution of a random variable specifies, for a given probability, the value that the random variable will fall below, with the specified probability. In fact it is an alternative to the probability density function (pdf).
Let X be a random variable with a distribution function F, and let \(\alpha \in (0,1)\). A value of x such that \(F(x)=P(X \le x) = \alpha \) is called a quantile of order \(\alpha \) for the distribution. Then, we can define the quantile function by:
Thus, the quantile function \(q_\alpha (X)\) yields the value that the random variable of the given distribution will fail to exceed, with probability \(\alpha \).
B Appendix: The Cornish–Fisher procedure
The Cornish–Fisher expansion is a useful tool for quantile estimations. For any \(\alpha \in (0,1)\), the upper \(\alpha \)th-quantile of \(F_n\) is defined by \(q_n(\alpha )={ inf} \left\{ x: F_n(x)\ge \alpha \right\} \), where \(F_n\) denotes the cumulative distribution function of \(\xi _n=(\sqrt{n}/\sigma )(\bar{X}-\mu )\), and \(\bar{X}\) is the sample mean of independent and identically distributed observations \(X_1,\ldots ,X_n\). If \(z_\alpha \) denotes the upper \(\alpha \)th-quantile of N(0, 1), then the fourth-order Cornish–Fisher expansion can be expressed as follows:
where S and K are the skewness and kurtosis of the observations \(X_i\), respectively.
The Cornish–Fisher expansion is useful because it allows one to obtain more accurate results compared to those acquired using the central limit theorem (CLT) approximation, which is the same as \(z_\alpha \) defined in the main text. A demonstration and example of the greater accuracy provided by the Cornish–Fisher expansion compared to the CLT approximation is reported by Chernozhukov et al. (2010).
In general, relation (23) grants a non-monotonic character to \(q_n(\alpha )\), which means that the true distribution’s ordering of quantiles is not preserved. Thus, the Cornish–Fisher expansion formula is valid only if the skewness and kurtosis coefficients of the distribution meet a particular constraint. This domain of validity has been studied by Maillard (2012), among others. Monotonicity requires the derivative of \(z_{{ CF},\alpha }\), relative to \(z_\alpha \), to be non-negative. This leads to the following constraint, which implicitly defines the domain of validity (D) of the Cornish–Fisher expansion:
In practice, this constraint is rarely considered, because S and K are generally considered to be small in finance applications.
C Online appendix: Quality of the estimation results for case 2 to case 5
Rights and permissions
About this article
Cite this article
Amédée-Manesme, CO., Barthélémy, F. & Maillard, D. Computation of the corrected Cornish–Fisher expansion using the response surface methodology: application to VaR and CVaR. Ann Oper Res 281, 423–453 (2019). https://doi.org/10.1007/s10479-018-2792-4
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10479-018-2792-4