Abstract
Lévy processes have become very popular in many applications in finance, physics and beyond. The Student–Lévy process is one interesting special case where increments are heavy-tailed and, for 1-increments, Student t distributed. Although theoretically available, there is a lack of path simulation techniques in the literature due to its complicated form. In this paper we address this issue using series representations with the inverse Lévy measure method and the rejection method and prove upper bounds for the mean squared approximation error. In the numerical section we discuss a numerical inversion scheme to find the inverse Lévy measure efficiently. We extend the existing numerical inverse Lévy measure method to incorporate explosive Lévy tail measures. Monte Carlo studies verify the error bounds and the effectiveness of the simulation routine. As a side result we obtain series representations of the so called inverse gamma subordinator which are used to generate paths in this model.
Similar content being viewed by others
References
Asmussen S, Rosiński J (2001) Approximations of small jumps of Lévy processes with a view towards simulation. J Appl Probab 38(2):482–493
Barndorff-Nielsen OE, Shephard N (2001) Non-Gaussian Ornstein–Uhlenbeck-based models and some of their uses in financial economics. J R Stat Soc Ser B (Stat Methodol) 63(2):167–241
Barndorff-Nielsen O, Halgreen C (1977) Infinite divisibility of the hyperbolic and generalized inverse Gaussian distributions. Zeitschrift für Wahrscheinlichkeitstheorie und Verwandte Gebiete 38(4):309–311
Blattberg RC, Gonedes NJ (1974) A comparison of the stable and Student distributions as statistical models for stock prices. J Bus 47(2):244–280
Bondesson L (1982) On simulation from infinitely divisible distributions. Adv Appl Probab 14(4):855–869
Bouchaud J-P, Potters M (2003) Theory of financial risk and derivative pricing: from statistical physics to risk management. Cambridge University Press, New York
Cassidy DT (2011) Describing n-day returns with Student’s t-distributions. Phys A 390(15):2794–2802
Cassidy DT, Hamp MJ, Ouyed R (2010) Pricing European options with a log Student’s t-distribution: a Gosset formula. Phys A 389(24):5736–5748
Cohen S, Rosiński J (2007) Gaussian approximation of multivariate Lévy processes with applications to simulation of tempered stable processes. Bernoulli 13(1):195–210
Cufaro Petroni N (2007) Mixtures in nonstable Lévy processes. J Phys A Math Theor 40(10):2227
Cufaro Petroni N, De Martino S, De Siena S, Illuminati F (2005) Lévy-Student distributions for halos in accelerator beams. Phys Rev E 72:066502
Derflinger G, Hörmann W, Leydold J (2010) Random variate generation by numerical inversion when only the density is known. ACM Trans Model Comput Simul 20(4):18:1–18:25
Devroye L (1981) On the computer generation of random variables with a given characteristic function. Comput Math Appl 7(6):547–552
Ferguson TS, Klass MJ (1972) A representation of independent increment processes without Gaussian components. Ann Math Stat 43(5):1634–1643
Grigelionis B (2012) Student’s t-distribution and related stochastic processes, Springer briefs in statistics. Springer, Berlin
Grosswald E (1976) The Student t-distribution of any degree of freedom is infinitely divisible. Zeitschrift für Wahrscheinlichkeitstheorie und Verwandte Gebiete 36(2):103–109
Grothe O, Schmidt R (2010) Scaling of Lévy Student processes. Phys A 389(7):1455–1463
Guo B-N, Qi F, Zhao J-L, Luo Q-M (2015) Sharp inequalities for polygamma functions. Math Slov 65(1):103–120
Heyde CC, Leonenko NN (2005) Student processes. Adv Appl Probab 37(2):342–365
Hilber N, Reich N, Schwab C, Winter C (2009) Numerical methods for Lévy processes. Finance Stoch 13(4):471–500
Hörmann W, Leydold J, Derflinger G (2004) Automatic nonuniform random variate generation, statistics and computing. Springer, Berlin
Hubalek F (2005) On the simulation from the marginal distribution of a Student t and generalized hyperbolic Lévy process, Working paper. https://pdfs.semanticscholar.org/4368/3935c410951d8145211a3d79148151cb07d8.pdf. Accessed 14 May 2018
Imai J, Kawai R (2011) On finite truncation of infinite shot noise series representation of tempered stable laws. Phys A 390(23):4411–4425
Imai J, Kawai R (2013) Numerical inverse Lévy measure method for infinite shot noise series representation. J Comput Appl Math 253:264–283
Massing T (2018) Local asymptotic normality for Student-Lévy processes under high-frequency sampling, Working paper. https://www.oek.wiwi.uni-due.de/fileadmin/fileupload/VWL-OEK/dokumente/LANHrB-Student-Levy.pdfHrB. Accessed 14 May 2018
Piessens R, Branders M (1974) A note on the optimal addition of abscissas to quadrature formulas of Gauss and Lobatto type. Math Comput 28(125):135–139
Rosiński J (1990) On series representations of infinitely divisible random vectors. Ann Probab 18(1):405–430
Rosiński J (2001) Series representations of Lévy processes from the perspective of point processes. In: Barndorff-Nielsen OE, Mikosch T, Resnick SI (eds) Lévy prcesses: theory and applications. Birkhäuser Boston, Boston, pp 401–415
Sato K-I (1999) Lévy processes and infinitely divisible distributions, Cambridge studies in advanced mathematics. Cambridge University Press, Cambridge
Schafheitlin P (1906) Die Lage der Nullstellen der Besselschen Funktionen zweiter Art. Sitzungsber Berlin Math Gesellschaft 5:82–93
Tankov P, Cont R (2015) Financial modelling with jump processes, second edition, Chapman and Hall/CRC financial mathematics series. Taylor & Francis, London
Todorov V, Tauchen G (2006) Simulation methods for Lévy-driven continuous-time autoregressive moving average (CARMA) stochastic volatility models. J Bus Econ Stat 24(4):455–469
Watson G (1995) A treatise on the theory of bessel functions, Cambridge mathematical library. Cambridge University Press, New York
Acknowledgements
The author is grateful to Christoph Hanck and Yannick Hoga as well as the anonymous reviewers for valuable comments which helped to substantially improve this paper and Friedrich Hubalek for graciously sharing his results. I thank Martin Arnold for excellent research assistance. Last but not least, I thank Theresa Kemper for carefully reading my work. Full responsibility is taken for all remaining errors.
Author information
Authors and Affiliations
Corresponding author
Electronic supplementary material
Below is the link to the electronic supplementary material.
A Appendix
A Appendix
In the “Appendix” we prove the statements of Sect. 3.
Proof of Lemma 2
Recall that
We use the inequality
first derived by Schafheitlin (1906); an elegant proof can be found in Watson (1995). Hence,
By standard integration,
such that (16) follows. To derive (17) the tail mass function is bounded by
Since \(Q([z,\infty ))\) is strictly decreasing and continuous, \(Q^{\leftarrow }(y)\), is the true inverse and
which completes the proof. \(\square \)
Proof of Theorem 2
Note that
Denote by \(q(y):=\frac{2\nu }{\pi y^2}\) the bound for \(Q^{\leftarrow }\) derived in Lemma 2. We start by proving (19). Taking expectations to obtain
where we have used the monotonicity of the expected value. Since the \(\Gamma _i\)s are \(\Gamma (i,1)\) distributed (with density function denoted by \(\gamma _i(x)\)), (33) is equal to
It remains to prove (20). Using the monotone convergence theorem
since \(Q^{\leftarrow }\left( \frac{\Gamma _i}{T}\right) {\mathbb {1}}_{\{U_i\le t\}}\ge 0\) for all i. Next, by the Cauchy–Schwarz inequality
Since \(i,j\ge n+1\ge 5\), we can bound (35) using \((i-1)(i-2)(i-3)(i-4)\ge (i-4)^4\) by
where \(\psi '(x)\) denotes the first derivative of the digamma function \(\psi (x):=\frac{\Gamma '(x)}{\Gamma (x)}\) (also called polygamma function of order 1). Guo et al. (2015) provide a sharp bound for polygamma functions. The inequality for \(\psi '(x)\) is
Applying (37) to (36) we obtain the bound
\(\square \)
Proof of Theorem 3
Again,
Note that \(E[V_i]=0\) and that \(V_i\) is independent of \(Q^{\leftarrow }\left( \frac{\Gamma _i}{T}\right) \) and \(U_i\). Hence, by Fubini’s theorem,
Furthermore, analogously to Theorem 2,
Since \(E[V_iV_j]=\delta _{i,j}\), (39) equals
as in the proof of Theorem 2. \(\square \)
Proof of Corollary 2
Since \(N_{\tau }=\#\{i\in {\mathbb {N}}:\Gamma _i\le \tau \}\) and the \(\Gamma _i\) are unit Poisson arrival times, \(N_{\tau }\sim Poi(\tau )\). We now use the law of iterated expectation.
by Theorem 3. The conditional expected value \(E\left[ \frac{1}{N_{\tau }-1}|\Gamma _2\le \tau \right] \) exists and \(N_{\tau }|\Gamma _2\le \tau \) follows a truncated Poisson distributed with density function
Hence, the conditional expectation is
which completes the proof. \(\square \)
Proof of Corollary 3
We only prove the claim for the deterministic truncation; the random truncation bound follows as in Corollary 2. Note that Lemma 2 implies that \(\frac{\mathrm {d}Q}{\mathrm {d}Q_0}\le 1\) and that the tail inverse \(Q_0^{\leftarrow }(y)=\frac{2\nu }{\pi y^2}\) exists in closed form. Let us start with the mean squared error
analogously as in the proof of Theorem 3, since \(E[V_iV_j]=\delta _{i,j}\). By the law of iterated expectation, this is equal to
because \(\frac{\mathrm {d}Q}{\mathrm {d}Q_0}\le 1\). The rest of the proof follows as in the proof of Theorem 3 since \(Q_0^{\leftarrow }\equiv q\). \(\square \)
Proof of Proposition 3
Asmussen and Rosiński (2001) showed that the distributional convergence is implied by
We show that \(\lim _{\varepsilon \rightarrow 0}\frac{\sigma _{\varepsilon }^2}{\varepsilon ^2}=+\infty \). Recall that
Using l’Hôspital’s rule, (41) is equal to
The monotone convergence theorem can be applied to (41) and thus
\(\square \)
Proof of Proposition 4
Note that in the non-finite variation case \(\mu _{\varepsilon }\) has to be zero (Sato 1999). Recall the Lévy measure for the univariate Student–Lévy process (with no drift and standard scaling) is given by
In the following we use the identity
for \(z>0\). As \(\Pi \) is a symmetric measure,
Again using l’Hôspital’s rule, for some constant \(C>0\) that may change from line to line
The second-to-last last step uses the monotone convergence theorem.
Rights and permissions
About this article
Cite this article
Massing, T. Simulation of Student–Lévy processes using series representations. Comput Stat 33, 1649–1685 (2018). https://doi.org/10.1007/s00180-018-0814-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00180-018-0814-y