Abstract
We propose and analyze a novel multi-index Monte Carlo (MIMC) method for weak approximation of stochastic models that are described in terms of differential equations either driven by random measures or with random coefficients. The MIMC method is both a stochastic version of the combination technique introduced by Zenger, Griebel and collaborators and an extension of the multilevel Monte Carlo (MLMC) method first described by Heinrich and Giles. Inspired by Giles’s seminal work, we use in MIMC high-order mixed differences instead of using first-order differences as in MLMC to reduce the variance of the hierarchical differences dramatically. This in turn yields new and improved complexity results, which are natural generalizations of Giles’s MLMC analysis and which increase the domain of the problem parameters for which we achieve the optimal convergence, \({\mathcal {O}}(\mathrm {TOL}^{-2}).\) Moreover, in MIMC, the rate of increase of required memory with respect to \(\mathrm {TOL}\) is independent of the number of directions up to a logarithmic term which allows far more accurate solutions to be calculated for higher dimensions than what is possible when using MLMC. We motivate the setting of MIMC by first focusing on a simple full tensor index set. We then propose a systematic construction of optimal sets of indices for MIMC based on properly defined profits that in turn depend on the average cost per sample and the corresponding weak error and variance. Under standard assumptions on the convergence rates of the weak error, variance and work per sample, the optimal index set turns out to be the total degree type. In some cases, using optimal index sets, MIMC achieves a better rate for the computational complexity than the corresponding rate when using full tensor index sets. We also show the asymptotic normality of the statistical error in the resulting MIMC estimator and justify in this way our error estimate, which allows both the required accuracy and the confidence level in our computational results to be prescribed. Finally, we include numerical experiments involving a partial differential equation posed in three spatial dimensions and with random coefficients to substantiate the analysis and illustrate the corresponding computational savings of MIMC.
Similar content being viewed by others
References
Amestoy, P.R., Duff, I.S., L’Excellent, J.-Y., Koster, J.: A fully asynchronous multifrontal solver using distributed dynamic scheduling. SIAM J. Matrix Anal. Appl. 23, 15–41 (2001)
Amestoy, P.R., Guermouche, A., L’Excellent, J.-Y., Pralet, S.: Hybrid scheduling for the parallel solution of linear systems. Parallel Comput. 32, 136–156 (2006)
Babuška, I., Nobile, F., Tempone, R.: A stochastic collocation method for elliptic partial differential equations with random input data. SIAM Rev. 52, 317–355 (2010)
Barth, A., Schwab, C., Zollinger, N.: Multi-level Monte Carlo finite element method for elliptic PDEs with stochastic coefficients. Numer. Math. 119, 123–161 (2011)
Bungartz, H., Griebel, M., Röschke, D., Zenger, C.: A proof of convergence for the combination technique for the Laplace equation using tools of symbolic computation. Math. Comput. Simul. 42, 595–605 (1996). (symbolic computation, new trends and developments, Lille 1993)
Bungartz, H.-J., Griebel, M.: Sparse grids. Acta Numer. 13, 147–269 (2004)
Bungartz, H.-J., Griebel, M., Röschke, D., Zenger, C.: Pointwise convergence of the combination technique for the Laplace equation. East-West J. Numer. Math. 2, 21–45 (1994)
Charrier, J., Scheichl, R., Teckentrup, A.: Finite element error analysis of elliptic PDEs with random coefficients and its application to multilevel Monte Carlo methods. SIAM J. Numer. Anal. 51, 322–352 (2013)
Cliffe, K., Giles, M., Scheichl, R., Teckentrup, A.: Multilevel Monte Carlo methods and applications to elliptic PDEs with random coefficients. Comput. Vis. Sci. 14, 3–15 (2011)
Collier, N., Haji-Ali, A.-L., Nobile, F., von Schwerin, E., Tempone, R.: A continuation multilevel Monte Carlo algorithm. BIT Numer. Math. 55, 399–432 (2015)
Durrett, R.: Probability: Theory and Examples, 2nd edn. Duxbury Press, Belmont (1996)
Giles, M.: Multilevel Monte Carlo path simulation. Oper. Res. 56, 607–617 (2008)
Giles, M., Reisinger, C.: Stochastic finite differences and multilevel Monte Carlo for a class of SPDEs in finance. SIAM J. Financ. Math. 3, 572–592 (2012)
Giles, M., Szpruch, L.: Antithetic multilevel Monte Carlo estimation for multi-dimensional SDEs without Lévy area simulation. Ann. Appl. Probab. (2013/4) (to appear)
Griebel, M., Harbrecht, H.: On the convergence of the combination technique. In: Preprint No. 2013–07, Institute of Mathematics, University of Basel, Switzerland (2013)
Griebel, M., Schneider, M., Zenger, C.: A combination technique for the solution of sparse grid problems. In: Iterative Methods in Linear Algebra (Brussels, 1991), pp. 263–281. North-Holland, Amsterdam (1992)
Haji-Ali, A.-L., Nobile, F., von Schwerin, E., Tempone, R.: Optimization of mesh hierarchies in multilevel Monte Carlo samplers. In: Analysis and Computations, Stochastic Partial Differential Equations (2015)
Harbrecht, H., Peters, M., Siebenmorgen, M.: Multilevel accelerated quadrature for PDEs with log-normal distributed random coefficient. In: Preprint No. 2013-18, Institute of Mathematics, University of Basel, Switzerland (2013)
Hegland, M., Garcke, J., Challis, V.: The combination technique and some generalisations. Linear Algebra Appl. 420, 249–275 (2007)
Heinrich, S.: Monte Carlo complexity of global solution of integral equations. J. Complex. 14, 151–175 (1998)
Heinrich, S., Sindambiwe, E.: Monte Carlo complexity of parametric integration. J. Complex. 15, 317–341 (1999)
Hoel, H., Schwerin, E.V., Szepessy, A., Tempone, R.: Adaptive multilevel Monte Carlo simulation. In: Engquist, B., Runborg, O., Tsai, Y.-H. (eds.) Numerical Analysis of Multiscale Computations. Lecture Notes in Computational Science and Engineering, vol. 82, pp. 217–234. Springer, New York (2012)
Hoel, H., von Schwerin, E., Szepessy, A., Tempone, R.: Implementation and analysis of an adaptive multilevel Monte Carlo algorithm. Monte Carlo Methods Appl. 20, 141 (2014)
Kebaier, A.: Statistical Romberg extrapolation: a new variance reduction method and applications to options pricing. Ann. Appl. Probab. 14, 2681–2705 (2005)
Kuo, F.Y., Schwab, C., Sloan, I.H.: Quasi-Monte Carlo finite element methods for a class of elliptic partial differential equations with random coefficients. SIAM J. Numer. Anal. 50, 3351–3374 (2012)
Mishra, S., Schwab, C.: Sparse tensor multi-level Monte Carlo finite volume methods for hyperbolic conservation laws with random initial data. Math. Comput. 81, 1979–2018 (2012)
Moraes, A., Tempone, R., Vilanova, P.: Multilevel hybrid Chernoff Tau-leap. BIT Numer. Math. (2015)
Nobile, F., Tamellini, L., Tempone, R.: Convergence of quasi-optimal sparse grid approximation of Hilbert-valued functions: application to random elliptic PDEs. In: MATHICSE Technical Report 12.2014. École Polytechnique Fédérale de Lausanne (2014) (submitted)
Pflaum, C.: Convergence of the combination technique for second-order elliptic differential equations. SIAM J. Numer. Anal. 34, 2431–2455 (1997)
Pflaum, C., Zhou, A.: Error analysis of the combination technique. Numer. Math. 84, 327–350 (1999)
Teckentrup, A., Scheichl, R., Giles, M., Ullmann, E.: Further analysis of multilevel Monte Carlo methods for elliptic PDEs with random coefficients. Numer. Math. 125, 569–600 (2013)
Teckentrup, A.L., Jantsch, P., Webster, C.G., Gunzburger, M.: A multilevel stochastic collocation method for partial differential equations with random input data. arXiv:1404.2647 (arXiv preprint) (2014)
van Wyk, H.-W.: Multilevel sparse grid methods for elliptic partial differential equations with random coefficients. arXiv:1404.0963v3 (arXiv preprint) (2014)
Xia, Y., Giles, M.: Multilevel path simulation for jump-diffusion SDEs. In: Plaskota, L., Woźniakowski, H. (eds.) Monte Carlo and Quasi-Monte Carlo Methods 2010, pp. 695–708. Springer, New York (2012)
Zenger, C.: Sparse grids. In: Parallel algorithms for partial differential equations (Kiel, 1990), vol. 31, pp. 241–251. Notes on Numerical Fluid Mechanics. Vieweg, Braunschweig (1991)
Acknowledgments
Raúl Tempone is a member of the Special Research Initiative on Uncertainty Quantification (SRI-UQ), Division of Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) at King Abdullah University of Science and Technology (KAUST). The authors would like to recognize the support of KAUST AEA project “Predictability and Uncertainty Quantification for Models of Porous Media” and University of Texas at Austin AEA Round 3 “Uncertainty quantification for predictive modeling of the dissolution of porous and fractured media”. F. Nobile acknowledges the support of the Swiss National Science Foundation under the Project No. 140574 “Efficient numerical methods for flow and transport phenomena in heterogeneous random porous media”. The authors would also like to thank Prof. Mike Giles for his valuable comments on this work.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix A: Asymptotic normality of the MIMC estimator
Lemma 5.1
(Asymptotic Normality of the MIMC Estimator). Consider the MIMC estimator introduced in (2), \({\mathcal {A}}\), based on a set of multi indices, \({\mathcal {I}}(\mathrm {TOL})\), and given by
Assume that for \(1\le i\le d\) there exists \(0<L_i(\mathrm {TOL})\) such that
Denote \({Y_{{{{\varvec{\alpha }}}}}= |{ \Delta \fancyscript{S} }_{{{{\varvec{\alpha }}}}}- {\mathrm {E}\left[ { \Delta \fancyscript{S} }_{{{{\varvec{\alpha }}}}}\right] }|}\) and assume that the following inequalities
hold for strictly positive constants \(\rho , \{s_i, r_i\}_{i=1}^d, Q_S\) and \(Q_R\). Choose the number of samples on each level, \(M_{{{{\varvec{\alpha }}}}}(\mathrm {TOL})\), to satisfy, for strictly positive sequences \(\{\tilde{s}_i\}_{i=1}^d\) and \(\{H_{{{\varvec{\tau }}}}\}_{{{{\varvec{\tau }}}}\in {\mathcal {I}}(\mathrm {TOL})}\) and for all \({{{{\varvec{\alpha }}}}}\in {\mathcal {I}}(\mathrm {TOL})\),
Denote, for all \(1 \le i \le d,\)
and choose \(0 < c_i\) such that whenever \(0 < p_i\), the inequality \(c_i < \rho /p_i\) holds. Finally, if we take the quantities \(L_i(\mathrm {TOL})\) in (54) to be
then we have
where \(\Phi (z)\) is the normal cumulative distribution function of a standard normal random variable.
Proof
We prove this theorem by ensuring that the Lindeberg condition [11, Lindeberg–Feller Theorem, p. 114] (also restated in [10, Theorem A.1]) is satisfied. The condition becomes in this case
for all \(\epsilon > 0\). Below we make repeated use of the following identity for non-negative sequences \({\{a_{{{{\varvec{\alpha }}}}}\}}\) and \({\{b_{{{{\varvec{\alpha }}}}}\}}\) and \(q \ge 0\):
First, we use the Markov inequality to bound
Using (58) and substituting for the variance \({\mathrm {Var}\left[ {\mathcal {A}}\right] }\) where we denote \({\mathrm {Var}\left[ { \Delta \fancyscript{S} }_{{{{\varvec{\alpha }}}}}\right] } = {\mathrm {E}\left[ ( \Delta \mathcal S_{{{{\varvec{\alpha }}}}}- {\mathrm {E}\left[ \Delta \mathcal S_{{{{\varvec{\alpha }}}}}\right] } )^{2}\right] }\) by \(V_{{{{\varvec{\alpha }}}}}\), we find
Using the lower bound in (56) on the number of samples, \(M_{{{{\varvec{\alpha }}}}}\), and (58), again yields
Finally, using the bounds (55a) and (55b),
Next, define three sets of dimension indices:
Then, using (54) yields
To conclude, observe that if \(|\hat{I}_3| = 0\), then \(\lim _{\mathrm {TOL}\downarrow 0} F= 0\) for any choice of \(L_i\ge 0,\,1\le i\le d\). Similarly, if \(|\hat{I}_3| > 0\), since we assumed that \(c_i p_i < \rho \) holds for all \(i \in \hat{I}_3\), then \(\lim _{\mathrm {TOL}\downarrow 0} F= 0\). \(\square \)
Remark
The lower bound on the number of samples per index (56) mirrors choice (9), the latter being the optimal number of samples satisfying constraint (7). Specifically, \(H_{{{{\varvec{\alpha }}}}}= \sqrt{V_{{{{\varvec{\alpha }}}}}W_{{{{\varvec{\alpha }}}}}}\) and \(\tilde{s}_i = s_i\). Furthermore, notice that the previous Lemma bounds the growth of L from above, while Theorems 2.1 and 2.2 bound the value of L from below to satisfy the bias accuracy constraint.
Appendix B: Integrating an exponential over a simplex
Lemma 6.1
The following identity holds for any \(L > 0\) and \(a \in \mathbb {R}\):
Proof
Then, we prove, by induction on d and for \(b = aL\), the following identity:
First, for \(d=1\), we have
Next, assuming that the identity is true for \(d-1\), we prove it for d. Indeed, we have
Finally, the second equality in (59) follows by repeatedly integrating by parts. \(\square \)
Lemma 6.2
For \(a \in \mathbb {R}^d\), assume \(A = \max _{i=1,2\ldots d} a_i > 0\) and denote
Then, for any \(L > 0\), there exists an \(\epsilon > 0\) satisfying
such that the following inequality holds:
Here, the constant \(\mathfrak {C_W}({{\varvec{a}}})\) is given by
Proof
First, note that \({{\varvec{a}}} = A {{\varvec{1}}}\) for some scalar \(A > 0\) and \({{\varvec{1}}} = (1,1,\ldots , 1) \) if and only if \(\mathfrak {a}_1 = d\). Then Lemma 6.1 immediately gives
Otherwise, recall that
holds for any \(x>0, b>0\) and \(j\in \mathbb {N}\). Then, using Lemma 6.1 and (62), we can write
continuing
\(\square \)
Lemma 6.3
The following inequality holds for any \(L \ge 1\) and \({{\varvec{a}}} \in \mathbb {R}_{+}^d\):
where
and
Proof
First, note that \({{\varvec{a}}} = A {{\varvec{1}}}\) for some scalar \(A > 0\) and \({{\varvec{1}}} = (1,1,\ldots , 1)\) if and only if \(\mathfrak {a}_1 = d\). Then Lemma 6.1 immediately gives:
Otherwise, without loss of generality, assume that \(a_i \le a_j\) for all \(1\le i \le j\le d\). Then, again using Lemma 6.1, we can write
where
Now consider
Here, we can bound
Recall that \({\epsilon } > 0\) and bound, using (62) for \((L-t)^j\) with \(b = {\epsilon }/{2}\),
\(\square \)
Appendix C: List of definitions
In this section, for easier reference, we list definitions of notation that is used in multiple pages or sections throughout the current work
Rights and permissions
About this article
Cite this article
Haji-Ali, AL., Nobile, F. & Tempone, R. Multi-index Monte Carlo: when sparsity meets sampling. Numer. Math. 132, 767–806 (2016). https://doi.org/10.1007/s00211-015-0734-5
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00211-015-0734-5