### Tail index estimation for dependent data

Sidney Resnick and Catalin Stărică
Source: Ann. Appl. Probab. Volume 8, Number 4 (1998), 1156-1183.

#### Abstract

A popular estimator of the index of regular variation in heavy-tailed models is Hill's estimator. We discuss the consistency of Hill's estimator when it is applied to certain classes of heavy-tailed stationary processes. One class of processes discussed consists of processes which can be appropriately approximated by sequences of m-dependent random variables and special cases of our results show the consistency of Hill's estimator for (i) infinite moving averages with heavy-tail innovations, (ii) a simple stationary bilinear model driven by heavy-tail noise variables and (iii) solutions of stochastic difference equations of the form $$Y_t = A_tY_{t-1} + Z_t, -\infty < t < \infty$$ where ${(A_n, Z_n), \quad -\infty < n < \infty}$ are iid and the Z's have a regularly varying tail probabilities. Another class of problems where our methods work successfully are solutions of stochastic difference equations such as the ARCH process where the process cannot be successfully approximated by m-dependent random variables. A final class of models where Hill estimator consistency is proven by our tail empirical process methods is the class of hidden semi-Markov models.

First Page:
Primary Subjects: 60G70
Secondary Subjects: 60G10
Full-text: Open access

Permanent link to this document: http://projecteuclid.org/euclid.aoap/1028903376
Mathematical Reviews number (MathSciNet): MR1661160
Digital Object Identifier: doi:10.1214/aoap/1028903376
Zentralblatt MATH identifier: 0942.60037

### References

Brockwell, P. and Davis, R. (1991). Time Series: Theory and Methods, 2nd ed. Springer, New York.
Mathematical Reviews (MathSciNet): MR92d:62001
Cline, D. (1983). Estimation and linear prediction for regression, autoregression and ARMA with infinite variance data. Ph.D. dissertation, Dept. Statist., Colorado State Univ.
Davis, R. and Resnick, S. (1984). Tail estimates motivated by extreme value theory. Ann. Statist. 12 1467-1487.
Zentralblatt MATH: 0555.62035
Mathematical Reviews (MathSciNet): MR760700
Digital Object Identifier: doi:10.1214/aos/1176346804
Project Euclid: euclid.aos/1176346804
Davis, R. and Resnick, S. (1988). Extremes of moving averages of random variables from the domain of attraction of the double exponential distribution. Stochastic Process. Appl. 30 41-68.
Zentralblatt MATH: 0657.60028
Mathematical Reviews (MathSciNet): MR968165
Digital Object Identifier: doi:10.1016/0304-4149(88)90075-0
Davis, R. and Resnick, S. (1996). Limit theory for bilinear processes with heavy tailed noise. Ann. Appl. Prob. 6 1191-1210.
Zentralblatt MATH: 0879.60053
Mathematical Reviews (MathSciNet): MR1422982
Digital Object Identifier: doi:10.1214/aoap/1035463328
Project Euclid: euclid.aoap/1035463328
Engle, R. (1982). Autoregressive conditional heteroscedastic models with estimates of the variance of United Kingdom inflation. Econometrica 50 987-1007.
Mathematical Reviews (MathSciNet): MR83j:62158
Zentralblatt MATH: 0491.62099
Digital Object Identifier: doi:10.2307/1912773
Feigin, P. and Resnick, S. (1996). Pitfalls of fitting autoregressive models for heavy-tailed time series. Unpublished manuscript.
Geluk, J., de Haan, L., Resnick, S. and St aric a, C. (1997). Second order regular variation, convolution and the central limit theorem. Stochastic Process. Appl. 69 139-159.
Mathematical Reviews (MathSciNet): MR98k:60081
Zentralblatt MATH: 0913.60001
Digital Object Identifier: doi:10.1016/S0304-4149(97)00042-2
Goldie, C. M. (1991). Implicit renewal theory and tails of solutions of random equations. Ann. Appl. Probab. 1 126-166.
Mathematical Reviews (MathSciNet): MR93i:60118
Zentralblatt MATH: 0724.60076
Digital Object Identifier: doi:10.1214/aoap/1177005985
Project Euclid: euclid.aoap/1177005985
Granger, C. W. J. and Andersen, A. (1978). Non-linear time series modelling. In Applied Time Series Analy sis (D. Findley, ed.) 25-38. Academic Press, New York.
Mathematical Reviews (MathSciNet): MR80d:62082
Zentralblatt MATH: 0465.62085
Grincevicius, A. (1975). One limit distribution for a random walk on the line. Lithuanian Math. J. 15 580-589.
Mathematical Reviews (MathSciNet): MR56:6877
de Haan, L. and Resnick, S. I. (1998). On asy mptotic normality of the Hill estimator. Stochastic Models. To appear.
Mathematical Reviews (MathSciNet): MR99e:62046
Digital Object Identifier: doi:10.1080/15326349808807504
de Haan, L., Resnick, S. I., Rootzen, H. and de Vries, C. (1989). Extremal behavior of solutions to a stochastic difference equation with applications to ARCH processes. Stochastic Process. Appl. 32 213-224.
Mathematical Reviews (MathSciNet): MR1014450
Zentralblatt MATH: 0679.60029
Digital Object Identifier: doi:10.1016/0304-4149(89)90076-8
Hall, P. (1982). On some simple estimates of an exponent of regular variation. J. Roy. Statist. Soc. Ser. B 44 37-42.
Zentralblatt MATH: 0521.62024
Mathematical Reviews (MathSciNet): MR655370
H¨ausler, E. and Teugels, J. (1985). On the asy mptotic normality of Hill's estimator for the exponent of regular variation. Ann. Statist. 13 743-756.
Hill, B. (1975). A simple approach to inference about the tail of a distribution. Ann. Statist. 3 1163-1174.
Zentralblatt MATH: 0323.62033
Mathematical Reviews (MathSciNet): MR378204
Digital Object Identifier: doi:10.1214/aos/1176343247
Project Euclid: euclid.aos/1176343247
Hsing, T. (1991). On tail estimation using dependent data. Ann. Statist. 19 1547-1569.
Mathematical Reviews (MathSciNet): MR92j:62035
Zentralblatt MATH: 0738.62026
Digital Object Identifier: doi:10.1214/aos/1176348261
Project Euclid: euclid.aos/1176348261
Hsing, T., H ¨usler, J. and Leadbetter, M. R. (1988). On the exceedance point process for a stationary sequence. Probab. Theory Related Fields 78 97-112.
Mathematical Reviews (MathSciNet): MR90d:60023
Zentralblatt MATH: 0619.60054
Digital Object Identifier: doi:10.1007/BF00718038
Kallenberg, O. (1983). Random Measures, 3rd ed. Akademie, Berlin.
Kesten, H. (1973). Random difference equations and renewal theory for products of random matrices. Acta Math. 131 207-248.
Mathematical Reviews (MathSciNet): MR55:13595
Zentralblatt MATH: 0291.60029
Digital Object Identifier: doi:10.1007/BF02392040
Leadbetter, M., Lindgren, G. and Rootzen, H. (1983). Extremes and Related Properties of Random Sequences and Processes. Springer, New York.
Mathematical Reviews (MathSciNet): MR84h:60050
Zentralblatt MATH: 0518.60021
Mason, D. (1982). Laws of large numbers for sums of extreme values. Ann. Probab. 10 754-764.
Zentralblatt MATH: 0493.60039
Mathematical Reviews (MathSciNet): MR659544
Digital Object Identifier: doi:10.1214/aop/1176993783
Project Euclid: euclid.aop/1176993783
Mason, D. (1988). A strong invariance theorem for the tail empirical process. Ann. Inst. H. Poincar´e Probab. Statist. 24 491-506.
Zentralblatt MATH: 0664.60038
Mason, D. and Turova, T. (1994). Weak convergence of the Hill estimator process. In Extreme Value Theory and Applications (J. Galambos, J. Lechner and Simiu, eds.) 419-431. Kluwer, Dordrecht.
Meier-Hellstern, K. S., Wirth, P. E., Yan, Y. L. and Hoeflin, D. A. (1991). Traffic models for ISDN data users: office automation application. In Teletraffic and Datatraffic in a Period of Change. Proceedings of the 13th ITC (A. Jensen and V. B. Iversen, eds.) 167-192. North-Holland, Amsterdam.
Mohler, R. R. (1973). Bilinear control processes with applications to engineering, ecology, and medicine. Math. Sci. Engrg. 106.
Mathematical Reviews (MathSciNet): MR48:10576
Zentralblatt MATH: 0343.93001
Resnick, S. (1987). Extreme Values, Regular Variation, and Point Processes. Springer, New York.
Mathematical Reviews (MathSciNet): MR89b:60241
Resnick, S. (1998). Why non-linearities can ruin the heavy tailed modeler's day. In A Practical Guide to Heavy Tails: Statistical Techniques for Analy sing Heavy Tailed Distributions (R. Adler, R. Feldman, M. S. Taqqu, eds.) Birkh¨auser, Boston. In press.
Mathematical Reviews (MathSciNet): MR1652288
Zentralblatt MATH: 0954.62107
Resnick, S. and St aric a, C. (1995). Consistency of Hill's estimator for dependent data. J. Appl. Probab. 32, 139-167. Resnick, S. and St aric a, C. (1997a). Smoothing the Hill estimator. Adv. in Appl. Probab. 29 271-293. Resnick, S. and St aric a, C. (1997b). Asy mptotic behavior of Hill's estimator for autoregressive data. Stochastic Models 13 703-723.
Mathematical Reviews (MathSciNet): MR96a:60021
Zentralblatt MATH: 0836.60020
Digital Object Identifier: doi:10.2307/3214926
Resnick, S. and Willekens, E. (1990). Moving averages with random coefficients and random coefficient autoregressive models. Stochastic Models 7 511-526.
Mathematical Reviews (MathSciNet): MR1139067
Zentralblatt MATH: 0747.60062
Digital Object Identifier: doi:10.1080/15326349108807204
Rootzen, H. (1995). The tail empirical process for stationary sequences. Preprint 1995:9 ISSN 1100-2255, Studies in Statistical Quality Control and Reliability, Chalmers Univ. Technology.
Rootzen, H., Leadbetter, M. and de Haan, L. (1990). Tail and quantile estimation for strongly mixing stationary sequences. Technical Report 292, Center for Stochastic Processes, Dept. Statistics, Univ. North Carolina, Chapel Hill, NC.
Vervaat, W. (1979). On a stochastic difference equation and a representation of non-negative infinitely divisible random variables. Adv. in Appl. Probab. 11 750-783.
Mathematical Reviews (MathSciNet): MR81b:60064
Digital Object Identifier: doi:10.2307/1426858