SWIFT Calibration of the Heston Model
Abstract
:1. Introduction
- We extend the SWIFT method to the calibration problem by deriving the option price gradient;
- We implement and test the speeding-up techniques mentioned in [1] based on multiple strike valuation;
- We propose a novel method for calibrating the Heston model with a set of options with certain fixed strikes that can be later used for arbitrary strikes by interpolation;
- We develop and implement speeding up techniques for the option price gradient.
1.1. Option Valuation
2. Heston Dynamics and Calibration Issues
- controls the position of the volatility surface;
- controls its skewness;
- and control the convexity of the surface;
- controls the term structure of implied volatility.
2.1. Calibration Challenges
- Heuristic based models. Using the relationships outlined above, some works reduce the dimension of the optimization problem by assuming some values or relationships between the parameters from the observation of a specific volatility surface. For example, [8] sets to the short-term ATM implied variance, obtained by using a BS model, a heuristic further justified by [11], where the linearity between and the BS implied volatility was verified for short maturities (less than 2 months). Other heuristics used in the industry are and setting to the BS short-term volatility [7]. These assumptions may restrict the optimization problem domain and exclude the optimum;
- Stochastic methods. They are typically used in combination with deterministic search methods, such as the Nelder–Mead simplex method [14] and avoid the pitfalls of the gradient-based methods if the optimization problem is not convex. Some examples are used in [11], and differential evolution and particle swar are used in [15]. These methods are too computationally expensive for real-time use as [16], which employs GPU computations to calibrate options using an SV model called SABR, and it took 421.72 s to calibrate 12 instruments with a tolerance of using 2 NVIDIA Geforce GTX470 GPUs.
2.2. The Characteristic Function
3. European Option Valuation and Calibration with SWIFT
3.1. Multi-Resolution Analysis and Shannon Wavelets
3.2. SWIFT Method
- Wavelet projection: the function f is replaced by its Shannon wavelet projection at scale ,
- Series truncation: the set of values of k involved in the sum of expression (14) is reduced to a finite interval ,It is important to notice that the first approximation lets us expressRemark 4.From this point onward, a symmetric interval will be considered for both convenience and consistency with the code implementation.
- Density coefficients approximation: the integral required to compute is replaced by an approximation , as will be shown in Section 3.2.1We can then define , and substitute into expression (1), obtainingFor European options, one can instead express it in terms of the strike-free payoff by defining , obtaining
- Payoff coefficients approximation: the integral required to compute is approximated in an analogous way as the integral to compute the density coefficients, and is replaced by an approximation , as will be shown in Section 3.2.1These coefficients can be precomputed when initializing the SWIFT procedure and shared across different strikes and maturities, saving computation time.
3.2.1. Density and Payoff Coefficients Approximation
3.3. Speeding Up the SWIFT Method
3.3.1. Fast Computation of the Density and Payoff Coefficients
3.3.2. Valuation with Multiple Strikes
- For each of the M strikes:
- −
- 1 FFT of length to compute density coefficients;
- −
- evaluations of the ChF ;
- 1 FFT of length to compute payoff coefficients;
- evaluations of the strike-free payoff integral defined in expression (21).
- For each of the M strikes:
- −
- evaluations of the x-dependent term of the ChF ;
- evaluations of the x-independent ChF ;
- 2 FFT of lengths and to compute the values of ;
- evaluations of the strike-free payoff integral defined in expression (21).
3.3.3. Fixed Set of Strikes
3.3.4. Option Price Gradient
- The value of can be reused for the price as well as for the gradient computations;
- If the parameters of the SWIFT method, are not changed during the gradient descent used in the calibration problem, then the values of both and can be reused throughout all the calibration steps;
- We can reuse the values of from the price computation to compute the gradient.
4. Calibration
5. Numerical Results
- Stress tests: the CP and SWIFT methods will be tested with several combinations of extreme strikes (ATM and deep ITM and OTM) as well as with long-term and short-term maturities, to detect any possible limitation or numerical issue in a wide usage range;
- Speed (The computations were performed on a 64-bit Ubuntu 18.04.4 LTS with a 3.70GHz Intel Core i7-8700K processor and 62.8 gigabytes of ram.) tests: the option calibration speeds for the regular SWIFT method (defined by expression (19)) and the one devised to quickly compute several option prices with different strike and the same maturity (defined by expression (29), which will be denoted KSWIFT), will be compared against CP for three different strike and expiry sets to check whether the multiple-strike alternative formulation is necessary to obtain a competitive option calibration method. These scenarios will represent:
- −
- A single expiry and multiple strikes;
- −
- A fixed number of maturities and a fixed number of strikes per maturity;
- −
- Different expiries for each strike.
When computing options with more than one different strike, a combination of OTM and ITM options will be used to provide an heterogeneous sample of contracts. Similarly, when more than one maturity is considered, a sample of long- and short-term expiries will be used;
5.1. Stress Tests
- CP and did not produce number (nan) results when evaluating very long expiries. Looking at the option price execution with the integrated debugger of the GDB compiler [33] showed that expression (9) runs into numerical overflow when the exponent of its hyperbolic functions is big enough (the same error happened when using the original expression provided in [20]). In most of the tests above, the overflow could be avoided when carefully setting an appropriate value for the upper bound of integral in expression (8), and by using a smaller value of the scale m. The error can also be avoided by selecting the ChF expression provided in [19] (we use this later on, and we denote the obtained prices by and present the results in Table 3);
- The SWIFT method at scale tends to underprice short expiry options. After checking the SWIFT parameters obtained through the parameter choice method defined in Section 3, it was observed that the initial value for , obtained by simply using the cumulant expression proposed in [1], resulted in a truncated Shannon wavelet expansion that did not cover a sufficient domain of the density function . A dynamical choice of the parameter based on the calculation of the area underneath the curve of the density function, as described in [1], can avoid this issue. Increasing the value of m also fixes the problem;
- None of the methods can handle the deep OTM option with a short expiry. The expected value should be close to but bigger than 0, as there are only 10 trading days to expiry and the price of the underlying should increase so that the option contract would not expire and become worthless. CP value seems too high and, in fact, when moving the value of in the interval , the price never clearly converges to a certain value, and it can give higher estimates for than 1.079 × 10−3, or even negative values. Changing the ChF expression does not fix this issue. SWIFT consistently gives it a price of 0. The contribution that makes the price different than zero probably lies in the tails of the distribution function, and one would require a really big value of c so that a point with a positive payoff is even considered in expression (29).
5.2. Speed Tests
- Generate option price values for each strike-expiry pair using as input. For this step, we used the SWIFT method (we also generated them using the CP method and checked that the difference between both results stayed under );
- Chose an initial guess for the calibration problem ;
- Solve the calibration problem with the desired method using as the initial guess and use the strike–expiry pairs and the prices obtained in the first step as inputs.
5.3. Realistic Convergence Tests
6. Conclusions and Future Research
- Most of the calibration tests with a single expiry have run into an optimal value different than the original one. It is to be seen if this is a property of the Heston model or if this was due instead to the specific parameter or strike/maturity values being used;
- It would be interesting to study the properties of the SWIFT implementation proposed for a chosen set of strikes in expression (32). We could interpolate the values at all strikes with spline methods that require derivatives, and not only derivative-free ones;
- Options with very long maturities may hamper the calibration process due to numerical overflows during the pricing step. The problem of long maturities has been tackled with Haar wavelets in [24]. It might be worth investigating whether we can do the same with Shannon wavelets;
- Deep OTM options with very short maturities are challenging to price. The problem seems to be the lack of accuracy of the approximation on the tails of the density function;
- Comparison with other calibration methods based on approximation formulae, like, for instance, the work by [34].
Author Contributions
Funding
Conflicts of Interest
Appendix A. Gradient Complimentary Formulas
Appendix B. Strike and Maturity Test Sets
Strike | Strike | Strike | Strike | Strike |
---|---|---|---|---|
0.9371 | 0.9956 | 1.0427 | 1.2287 | 1.3939 |
0.8603 | 0.9868 | 1.0463 | 1.2399 | 1.4102 |
0.8112 | 0.9728 | 1.0499 | 1.2485 | 1.4291 |
0.7760 | 0.9588 | 1.0530 | 1.2659 | 1.4456 |
0.7470 | 0.9464 | 1.0562 | 1.2646 | 1.4603 |
0.7216 | 0.9358 | 1.0593 | 1.2715 | 1.4736 |
0.6699 | 0.9175 | 1.0663 | 1.2859 | 1.5005 |
0.6137 | 0.9025 | 1.0766 | 1.3046 | 1.5328 |
Expiry | Strike | Strike | Strike | Strike | Strike |
---|---|---|---|---|---|
0.119047619047619 | 0.9371 | 0.9956 | 1.0427 | 1.2287 | 1.3939 |
0.238095238095238 | 0.8603 | 0.9868 | 1.0463 | 1.2399 | 1.4102 |
0.357142857142857 | 0.8112 | 0.9728 | 1.0499 | 1.2485 | 1.4291 |
0.476190476190476 | 0.7760 | 0.9588 | 1.0530 | 1.2659 | 1.4456 |
0.595238095238095 | 0.7470 | 0.9464 | 1.0562 | 1.2646 | 1.4603 |
0.714285714285714 | 0.7216 | 0.9358 | 1.0593 | 1.2715 | 1.4736 |
1.07142857142857 | 0.6699 | 0.9175 | 1.0663 | 1.2859 | 1.5005 |
1.42857142857143 | 0.6137 | 0.9025 | 1.0766 | 1.3046 | 1.5328 |
References
- Ortiz-Gracia, L.; Oosterlee, C.W. A highly efficient Shannon wavelet inverse fourier technique for pricing European options. SIAM J. Sci. Comput. 2016, 38, B118–B143. [Google Scholar] [CrossRef] [Green Version]
- Leitao, A.; Ortiz-Gracia, L.; Wagner, E.I. SWIFT valuation of discretely monitored arithmetic Asian options. J. Comput. Sci. 2018, 28, 120–139. [Google Scholar] [CrossRef]
- Maree, S.C.; Ortiz-Gracia, L.; Oosterlee, C.W. Pricing early-exercise and discrete barrier options by Shannon wavelet expansions. Numer. Math. 2017, 136, 1035–1070. [Google Scholar] [CrossRef]
- Moré, J.J. The Levenberg-Marquardt algorithm: Implementation and theory. In Numerical Analysis; Springer: Berlin/Heidelberg, Germany, 1978. [Google Scholar]
- Cui, Y.; del Baño Rollin, S.; Germano, G. Full and fast calibration of the Heston stochastic volatility model. Eur. J. Oper. Res. 2017, 263, 625–638. [Google Scholar] [CrossRef]
- Heston, S.L. A Closed-Form Solution for Options with Stochastic Volatility with Applications to Bond and Currency Options. Rev. Financ. Stud. 1993, 6, 327–343. [Google Scholar] [CrossRef] [Green Version]
- Clark, I. Foreign Exchange Option Pricing: A Practitioner’s Guide; Wiley: Chichester, UK, 2011. [Google Scholar]
- Gatheral, J. The volatility surface: A practitioner’s guide. In Finance; Wiley: Chichester, UK, 2006. [Google Scholar]
- Gilli, M.; Schumann, E. Calibrating option pricing models with heuristics. In Natural Computing in Computational Finance. Studies in Computational Intelligence; Brabazon, A., O’Neill, M., Maringer, D., Eds.; Springer: Berlin/Heidelberg, Germany, 2011; Volume 380. [Google Scholar]
- Janek, A.; Kluge, T.; Weron, R.; Wystup, U. FX smile in the Heston model. In Statistical Tools for Finance and Insurance; Cizek, P., Härdle, W., Weron, R., Eds.; Springer: Berlin/Heidelberg, Germany, 2011; pp. 133–162. [Google Scholar]
- Bin, C. Calibration of the Heston Model with Application in Derivative Pricing and Hedging. Master’s Thesis, TU Delft, Delft, The Netherlands, 2007. [Google Scholar]
- Mikhailov, S.; Nögel, U. Heston’s stochastic volatility model implementation, calibration and some extensions. Wilmott Mag. 2003, 4, 74–79. [Google Scholar]
- Gerlich, F.; Giese, A.M.; Maruhn, J.H.; Sachs, E.W. Parameter identification in financial market models with a feasible point SQP algorithm. Comput. Optim. Appl. 2012, 51, 1137–1161. [Google Scholar] [CrossRef]
- Lagarias, J.C.; Reeds, J.A.; Wright, M.H.; Wright, P.E. Convergence properties of the Nelder-Mead simplex method in low dimensions. SIAM J. Optim. 1998, 9, 112–147. [Google Scholar] [CrossRef] [Green Version]
- Gilli, M.; Schumann, E. Heuristic optimisation in financial modelling. Ann. Oper. Res. 2012, 193, 129–158. [Google Scholar] [CrossRef] [Green Version]
- Fernández, J.L.; Ferreiro, A.M.; García-Rodrxixguez, J.A.; Leitao, A.; López-Salas, J.G.; Vxaxzquez, C. Static and dynamic SABR stochastic volatility models: Calibration and option pricing using GPUs. Math. Comput. Simul. 2013, 94, 55–75. [Google Scholar] [CrossRef]
- Kahl, C.; Jäckel, P. Not-so-complex logarithms in the Heston model. Wilmott Mag. 2006, 19, 94–103. [Google Scholar]
- Albrecher, H.; Mayer, P.; Schoutens, W.; Tistaert, J. The little Heston trap. Wilmott 2007, 1, 83–92. [Google Scholar]
- Schoutens, W.; Simons, E.; Tistaert, J. A perfect calibration! now what? Wilmott 2004, 2, 66–78. [Google Scholar] [CrossRef]
- del Baño Rollin, S.; Ferreiro-Castilla, A.; Utzet, F. On the density of log-spot in the Heston volatility model. Stoch. Process. Their Appl. 2010, 120, 2037–2063. [Google Scholar] [CrossRef] [Green Version]
- Mallat, S. A Wavelet Tour of Signal Processing: The Sparse Way; Academic Press: Cambridge, MA, USA, 2009. [Google Scholar]
- Daubechies, I. Ten Lectures on Wavelets; CBMS-NSF Regional Conference Series in Applied Mathematics; Society for Industrial and Applied Mathematics: Philadelphia, PA, USA, 1992. [Google Scholar]
- Cattani, C. Shannon wavelets theory. Math. Probl. Eng. 2008, 2008, 164808. [Google Scholar] [CrossRef] [PubMed]
- Ortiz-Gracia, L.; Oosterlee, C.W. Robust pricing of European options with wavelets and the characteristic function. SIAM J. Sci. Comput. 2013, 35, B1055–B1084. [Google Scholar] [CrossRef] [Green Version]
- Fang, F.; Oosterlee, C.W. A novel pricing method for european options based on fourier-cosine series expansions. SIAM J. Sci. Comput. 2008, 31, 826–848. [Google Scholar] [CrossRef] [Green Version]
- Frigo, M.; Johnson, S.G. The design and implementation of FFTW3. Proc. IEEE 2005, 93, 216–231. [Google Scholar] [CrossRef] [Green Version]
- Floc’h, F.L. Notes on the SWIFT method based on Shannon Wavelets for Option Pricing. arXiv 2020, arXiv:2005.13252. [Google Scholar]
- Christoffersen, P.; Jacobs, K. The importance of the loss function in option valuation. J. Financ. Econ. 2004, 72, 291–318. [Google Scholar] [CrossRef] [Green Version]
- Lourakis, M.I.A. Levmar: Levenberg–Marquardt Nonlinear Least Squares Algorithms in C/C++. 2004. Available online: https://github.com/jturney/levmar (accessed on 2 March 2020).
- Planitz, M.; Anderson, E. LAPACK Users Guide. Math. Gaz. 1995, 79, 210. [Google Scholar] [CrossRef]
- Glasserman, P.; Kim, K.K. Gamma expansion of the Heston stochastic volatility model. Financ. Stochastics 2011, 15, 267–296. [Google Scholar] [CrossRef] [Green Version]
- Andersen, L. Simple and efficient simulation of the Heston stochastic volatility model. J. Comput. Financ. 2008, 11, 1–42. [Google Scholar] [CrossRef]
- Willmore, F.T. Debugging with gdb. In Introduction to Scientific and Technical Computing; CRC Press: Boca Raton, FL, USA, 2016. [Google Scholar]
- Alòs, E.; Santiago, R.D.; Vives, J. Calibration of stochastic volatility models via second-order approximation: The Heston case. Int. J. Theor. Appl. Financ. 2015, 18, 1550036. [Google Scholar] [CrossRef]
Name | |||||
---|---|---|---|---|---|
3 | 0.1 | 0.25 | −0.8 | 0.08 | |
1.5768 | 0.0398 | 0.0175 | −0.5711 | 0.0175 | |
0.5 | 0.04 | 1 | −0.9 | 0.04 | |
0.3 | 0.04 | 0.9 | −0.5 | 0.04 | |
1 | 0.09 | 1 | 0.04 | 0.09 |
Parameters | S | K | ||||
---|---|---|---|---|---|---|
100 | 50 | 45 | 65.565 | nan | nan | |
100 | 100 | 45 | 46.911 | nan | nan | |
100 | 200 | 45 | 27.198 | nan | nan | |
100 | 50 | 0.04 | 44.221 | 50.000 | 50.000 | |
100 | 100 | 0.04 | 0.380 | 1.045 | 1.046 | |
100 | 200 | 0.04 | 0 | 0 | 1.079 |
Parameters | S | K | ||||
---|---|---|---|---|---|---|
100 | 50 | 45 | 6 | 65.565 | 65.565 | |
100 | 100 | 45 | 6 | 46.911 | 46.911 | |
100 | 200 | 45 | 6 | 27.198 | 27.198 | |
100 | 50 | 0.04 | 200 | 50.000 | 50.000 | |
100 | 100 | 0.04 | 200 | 1.046 | 1.046 | |
100 | 200 | 0.04 | 300 | −1.174 | 1.079 |
Strike and Maturities Set | Heston Parameters | Method | Time (Seconds) | I | |
---|---|---|---|---|---|
Set 1 | SWIFT | 6.9 | 10 | 3.932 | |
Set 1 | KSWIFT | 4.5 | 10 | 3.932 | |
Set 1 | CP | 4.6 | 10 | 3.932 | |
Set 2 | SWIFT | 35.9 | 13 | 1.002 | |
Set 2 | KSWIFT | 5.0 | 13 | 1.002 | |
Set 2 | CP | 6.3 | 13 | 1.002 | |
Set 3 | KSWIFT | 1.7 | 13 | 1.002 |
6.640 | 2.657 | 1.160 | |
1.547 | 1.321 | 1.746 | |
1.978 | 2.248 | 3.725 | |
2.649 | 1.365 | 8.661 | |
3.629 | 4.790 | 8.339 | |
Iterations | 14 | 6 | 7 |
Time (seconds) | 3.3 | 1.9 | 2.0 |
2.867 | 2.030 | 3.643 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2021 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 (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Romo, E.; Ortiz-Gracia, L. SWIFT Calibration of the Heston Model. Mathematics 2021, 9, 529. https://doi.org/10.3390/math9050529
Romo E, Ortiz-Gracia L. SWIFT Calibration of the Heston Model. Mathematics. 2021; 9(5):529. https://doi.org/10.3390/math9050529
Chicago/Turabian StyleRomo, Eudald, and Luis Ortiz-Gracia. 2021. "SWIFT Calibration of the Heston Model" Mathematics 9, no. 5: 529. https://doi.org/10.3390/math9050529
APA StyleRomo, E., & Ortiz-Gracia, L. (2021). SWIFT Calibration of the Heston Model. Mathematics, 9(5), 529. https://doi.org/10.3390/math9050529