Abstract
In this paper we describe a novel approach to the study of U-statistics in the Markovian setup based on the (pseudo-) regenerative properties of Harris Markov chains. Exploiting the fact that any sample path X 1, ..., X n of a general Harris chain X may be divided into asymptotically i.i.d. data blocks \(\mathcal{B}_1 , \ldots \mathcal{B}_N\) of random length corresponding to successive (pseudo-) regeneration times, we introduce the notion of regenerative U-statistic Ω N = Σ k≠l ω h \(\left( {\mathcal{B}_k ,\mathcal{B}_l } \right)\)/(N(N − 1)) related to a U-statistic U n = Σ i≠j h(X i , X j )/(n(n − 1)). We show that, under mild conditions, these two statistics are asymptotically equivalent up to the order O ℙ(n−1). This result serves as a basis for establishing limit theorems related to statistics of the same form as U n . Beyond its use as a technical tool for proving results of a theoretical nature, the regenerative method is also employed here in a constructive fashion for estimating the limiting variance or the sampling distribution of certain U-statistics through resampling. The proof of the asymptotic validity of this statistical methodology is provided, together with an illustrative simulation result.
Similar content being viewed by others
References
J. Aaronson, R. Burton, H. G. Dehling, D. Gilat, T. Hill, and B. Weiss, “Strong Laws for L- and U-Statistics”, Trans. Amer. Math. Soc. 348, 2845–2866 (1996).
S. Asmussen, Applied Probability and Queues (Springer, New York, 2003).
K. B. Athreya and G. S. Atuncar, “Kernel Estimation for Real-Valued Markov Chains”, Sankhya 60(1), 1–17 (1998).
K. B. Athreya and P. Ney, “A New Approach to the Limit Theory of Recurrent Markov Chains”, Trans. Amer. Math. Soc. 245, 493–501 (1978).
K. B. Athreya, D. McDonald, and P. Ney, “Limit Theorems for Semi-Markov Processes and Renewal Theory for Markov Chains”, Ann. Probab. 6(5), 788–797 (1978).
N. G. Becker and S. Utev, “Threshold Results for U-Statistics of Dependent Binary Variables”, J. Theor. Probab. 14(1), 97–114 (2001).
W. Bednorz,K. Latuszynski, and R. Latala, “A RegenerationProof of the Central Limit Theorem forUniformly ErgodicMarkov Chains”, Elect. Comm. Probab. 13, 85–98 (2008).
P. Bertail and S. Clémençon, “Edgeworth Expansions for Suitably Normalized Sample Mean Statistics of Atomic Markov Chains”, Probab. Theory Rel. Fields 130(3), 388–414 (2004).
P. Bertail and S. Clémençon, “Regenerative-Block Bootstrap for Markov Chains”, Bernoulli 12(4) (2005).
P. Bertail and S. Clémençon, “Regeneration-Based Statistics for Harris Recurrent Markov Chains,” in Lecture Notes in Statist., Vol. 187: Probability and Statistics for Dependent Data, ed. by P. Doukhan, P. Bertail, and P. Soulier (Springer, 2006), pp. 1–54.
P. Bertail, S. Clémençon, and J. Tressou, “Extreme Value Statistics for Markov Chains via the (Pseudo-)Regenerative Method”, Extremes (2009) (in press). Preprint available at http://hal.archives-ouvertes.fr/hal-00165652.
P. Bertail and S. Clémençon, “Approximate Regenerative Block-Bootstrap for Markov Chains”, Comput. Statist. and Data Analysis 52(5), 2739–2756 (2007).
P. Bertail and S. Clémençon, “Sharp Bounds for the Tails of Functionals of Markov Chains”, Theory Probab. Appl. 53(3), 505–515 (2009).
P. Bertail, S. Clémençon, and J. Tressou, Regenerative Block-Bootstrap Confidence Intervals for the Extremal Index of Markov Chains”, in Intern. Workshop in Appl. Probab., 2008. Available at http://hal.archives-ouvertes.fr/hal-00214306/fr/.
E. Bolthausen, “The Berry-EsséenTheorem for Functionals of DiscreteMarkov Chains”, Z.Wahrsch. Verw. Geb. 54(1), 59–73 (1980).
E. Bolthausen, “The Berry-Esséen Theorem for Strongly Mixing Harris Recurrent Markov Chains”, Z. Wahrsch. Verw. Geb. 60(3), 283–289 (1982).
S. Borovkova, R. Burton, and H. G. Dehling, “Consistency of the Takens Estimator for the Correlation Dimension”, Ann. Appl. Probab. 9(2), 376–390 (1999).
S. Borovkova, R. Burton, and H. G. Dehling, “From Dimension Estimation to Asymptotics of Dependent U-Statistics”, in Limit Theorems in Probab. and Statist. I, ed. by I. Berkes, E. Csaki, and M. Csörgő (Budapest, 1999), pp. 201–234.
S. Borovkova, R. Burton, and H. G. Dehling, “Limit Theorems for Functionals of Mixing Processes with Applications to U-Statistics and Dimension Estimation”, Trans. Amer. Math. Soc. 353, 4261–4318 (2001).
H. Callaert and N. Veraverbeke, “The Order of the Normal Approximation for a Studentized U-Statistic”, Ann. Statist. 9(1), 194–200 (1981).
S. Clémençon, “Adaptive Estimation of the Transition Density of a Regular Markov Chain by Wavelet Methods”,Math. Methods Statist. 9(4), 323–357 (2000).
S. Clémençon, “Moment and Probability Inequalities for Sums of Bounded Additive Functionals of a Regular Markov Chains via the Nummelin Splitting Technique”, Statist. Probab. Lett. 55, 227–238 (2001).
H. G. Dehling, Limit Theorems for Dependent U-Statistics, in Lecture Notes in Statist. (2006), Vol. 187, pp. 65–86.
M. Denker and G. Keller, “On U-Statistics and von Mises Statistics for Weakly Dependent Processes”, Z. Wahrsch. Verw. Geb. 64, 505–522 (1983).
M. Denker and G. Keller, “Rigorous Statistical Procedures for Data from Dynamical Systems”, J. Statist. Phys. 44, 67–93, 1986.
W. Doeblin, “Sur deux problemes de M. Kolmogorov concernant les chaînes dénombrables”, Bull. Soc. Math. de France 66, 210–220 (1938).
W. Doeblin, “Eléments d’une théorie générale des chaînes simple constantes de Markov”, Ann. Sci. Ecole Norm. Sup. 37(3), 61–111 (1940).
R. Douc, A. Guillin, and E. Moulines, “Bounds on Regeneration Times and Limit Theorems for Subgeometric Markov Chains”, Ann. Inst. H. Poincaré-Probab. et Statist. 44(2), 239–257 (2008). 0246-0203.
P. Doukhan, Mixing: Properties and Examples, in Lecture Notes in Statist. (Springer, New York, 1984), Vol. 85.
J. Dubinskaite, “Limit Theorems in ℝk”, Lithuanian Math. J. 24, 256–265 (1984).
W. R. Gilks, S. Richardson, and D. J. Spiegelhalter, Markov Chain Monte Carlo in Practice (Chapman & Hall, 1996).
J. Hájek, “Asymptotically Most Powerful Rank Tests”, Ann. Math. Statist. 33, 1124–1147 (1962).
W. Hoeffding, “A Class of Statistics with Asymptotically Normal Distribution”, Ann. Statist. 19, 293–325 (1948).
J. L. Jensen, “Asymptotic Expansions for Strongly Mixing Harris Recurrent Markov Chains”, Scand. J. Statist. 16, 47–63 (1989).
V. V. Kalashnikov, Topics on Regenerative Processes (CRC Press, 1994).
S. N. Lahiri, Resampling Methods for Dependent Data (Springer, 2003).
V. K. Malinovskii, “On Limit Theorems for the Number of Markov Renewals”, in Lecture Notes in Math., B (Springer, 1985), Vol. 1155, pp. 190–222.
V. K. Malinovskii, “Second Order Efficiency of an Asymptotically Efficient Test Based on Markov Observations”, Theor. Probab. Appl. 30, 603–608 (1985).
V. K. Malinovskii, “On Some Asymptotic Relations and Identities for Harris Recurrent Markov Chains”, in Statistics and Control of Stochastic Processes (1985), pp. 317–336.
V. K. Malinovskii, “Limit Theorems for HarrisMarkov Chains. I”, Theory Probab. Appl. 31, 269–285 (1987).
V. K. Malinovskii, “Limit Theorems for Recurrent Semi-MarkovProcesses and Markov Renewal Processes”, J. Soviet Math. 36, 493–502 (1987).
V. K. Malinovskii, “On Integral and Local Limit Theorems for Recurrent Markov Renewal Processes”, J. Soviet Math. 57(4), 3286–3301 (1988).
V. K. Malinovskii, “On a Limit Theorem for Dependent Random Variables”, Ann. Acad. Sci. Fenicae, Ser. A.I. Mathematica 13, 225–229 (1988).
V. K. Malinovskii, “Limit Theorems for Harris Markov chains. II”, Theory Probab. Appl. 34, 252–265 (1989).
V. K. Malinovskii, “Asymptotic Optimality of Tests in a Hypothesis Testing Problem for Recurrent Jump Markov Processes”, J. Soviet Math. 44, 403–510 (1989).
V. K. Malinovskii, “Power Functions and the Defficiency of Asymptotically Efficient Tests in the Case of Markov observations”, Theor. Probab. Appl. 34, 441–453 (1989).
V. K. Malinovskii, “Large Deviations for Recurrent Markov Renewal Processes”, Theor. Probab. Appl. 36, 170–173 (1991).
V. K. Malinovskii, “Asymptotic Optimality of Criteria in the Problem of Testing Hypotheses for a Recurrent Semi-Markov Processes”, J. Soviet. Math. 59(4), 955–959 (1992).
V. K. Malinovskii, “Limit Theorems for Stopped Random Sequences. I. Rates of Convergence and Asymptotic Expansions”, Theor. Probab. Appl. 38, 673–693 (1993).
V. K. Malinovskii, “Asymptotic Expansions in the Sequential Estimation of an Autoregressive Parameter”, Math. Methods Statist. 2, 206–227 (1993).
V. K. Malinovskii, “Asymptotic Expansions in Sequential Estimation for the first Order Random Coefficient Autoregressive Model: Regenerative Approach”, Acta Applicandae Mathematica 34, 261–281 (1994).
V. K. Malinovskii, “Limit Theorems for Stopped Random Sequences. II. Probabilities of Large Deviations”, Theor. Probab. Appl. 41(1), 70–90 (1996).
S. P. Meyn and R. L. Tweedie,Markov Chains and Stochastic Stability (Springer, 1996).
E. Nummelin, “A splitting technique for Harris recurrent chains”, Z. Wahrsch. Verw. Geb. 43, 309–318 (1978).
E. Nummelin, “Uniform and Ratio Limit Theorems for Markov Renewal and Semi-Regenerative Processes on a General State Space”, Ann. Inst. H. Poincare, Sect. B. V.XIV(2), 119–143 (1978).
E. Nummelin, “General Irreducible Markov Chains and Non-Negative Operators”, Cambridge Tracts in Math. (1984).
V. V. Petrov, Limit Theorems of Probability Theory: Sequences of Independent Random Variables (Oxford Science Publ., 1995).
D. Revuz, Markov Chains, 2nd ed. (North-Holland, 1984).
E. Rio, Théorie asymptotique des processus al éatoires faiblement d épendants, in Mathématiques et Applications (Springer, 2000), Vol. 31.
C. P. Robert and R. Casella, Monte Carlo Statistical Methods (Springer, 1999).
R. J. Serfling, Approximation Theorems of Mathematical Statistics (Wiley, New York, 1980).
G. Shorack, Probability for Statisticians (Springer, 2000).
I.G. Shevtsova, “Sharpening of the Upper Bound of the Absolute Constant in the Berry-Esséen Inequality”, Theory Probab. Appl. 51(3), 549–553 (2007).
W. L. Smith, “Regenerative Stochastic Processes”, Proc. Roy. Statist. Soc. 232, 6–31 (1955).
H. Thorisson, Coupling, Stationarity, and Regeneration (Springer, New York, 2000).
P. Tuominen and R. L. Tweedie, “Subgeometric Rates of Convergence of f-Ergodic Markov Chains”, Adv. Appl. Probab. 26, 775–798 (1994).
K. Yoshihara, “Limiting Behavior of U-Statistics for Stationary Absolurely Regular Processes”, Z. Wahrsch. Verw. Geb. 35, 237–252 (1976).
Author information
Authors and Affiliations
Corresponding author
About this article
Cite this article
Bertail, P., Clémençon, S. A renewal approach to Markovian U-statistics. Math. Meth. Stat. 20, 79–105 (2011). https://doi.org/10.3103/S1066530711020013
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.3103/S1066530711020013
Keywords
- Markov chain
- regenerative process
- Nummelin splitting technique
- U-statistics
- Hoeffding decomposition
- limit theorems
- Berry-Esséen inequality
- regenerative block-bootstrap