On sequential estimation of parameters in semimartingale regression models with continuous time parameter. (English) Zbl 1043.62067

Galtchouk, L.; Konev, V.

Ann. Stat. 29, No. 5, 1508-1536 (2001).

Summary: We consider the problem of parameter estimation for multidimensional continuous-time linear stochastic regression models with an arbitrary finite number of unknown parameters and with martingale noise. The main result of the paper claims that the unknown parameters can be estimated with prescribed mean-square precision in this general model providing a unified description of both discrete and continuous time process. Among the conditions on the regressors there is one bounding the growth of the maximal eigenvalue of the design matrix with respect to its minimal eigenvalue. This condition is slightly stronger as compared with the corresponding conditions usually imposed on the regressors in asymptotic investigations but still it enables one to consider models with different behavior of the eigenvalues. The construction makes use of a two-step procedure based on the modified least-squares estimators and special stopping rules.

MSC:

62L12	Sequential estimation
62M10	Time series, auto-correlation, regression, etc. in statistics (GARCH)
60G48	Generalizations of martingales
62M09	Non-Markovian processes: estimation

Keywords:

weighted least squares; prescribed precision

References:

[1]	Anderson, T. W. and Taylor, J. B. (1976). Strong consistency of least squares estimates in normal linear regression. Ann. Statist. 4 788-790. · Zbl 0339.62039 · doi:10.1214/aos/1176343552
[2]	Aras, G. (1990). Sequential estimation of the mean of a first-order autoregressive process. Comm. Statist. Theory Methods 19 1639-1652. · Zbl 0724.62080 · doi:10.1080/03610929008830282
[3]	Arato, M. (1982). Linear stochastic systems with constant coefficients. A statistical approach. Lecture Notes in Control Inform. Sci. 45 Springer, New York. · Zbl 0544.93060
[4]	Arato, M., Kolmogorov, A. N. and Sinai, Y. G. (1962). On parameter estimation of a complex stationary gaussian process. Doklady Acad. USSR 146 747-750. · Zbl 0136.40702
[5]	Borisov, V. Z. and Konev V. V. (1977). On sequential parameter estimation in discrete time processes. Automat. Remote Control 38 58-64. · Zbl 0423.93046
[6]	Christopeit, N. (1986). Quasi-least-squares estimation in semimartingale regression models. Stochastics 16 255-278. · Zbl 0586.62137 · doi:10.1080/17442508608833376
[7]	Darwich, A. R. and Le Breton, A. (1991). About asymptotic behavior of multidimensional Gaussian martingales in normal linear regression. Statist. Probab. Lett. 12 317-321. · Zbl 0742.60047 · doi:10.1016/0167-7152(91)90099-D
[8]	Dellacherie, C. (1972). Capacites et processus stochastiques. Springer, New York. · Zbl 0246.60032
[9]	Galtchouk, L. I. (1975). The structure of a class of martingales. Proceedings of the Seminar on Random Processes 7-32. Academy of Sciences of Lithuania, Vilnus.
[10]	Galtchouk, L. I. (1976). Representation of some martingales. Theory Probab. Appl. 21 613-620.
[11]	Galtchouk, L. I. (1980). Optional martingales. Math. Sbornik 112 483-521. · Zbl 0449.60036
[12]	Galtchouk, L. I. and Maljutov, M.B. (1995). One bound for the mean duration of sequential testing homogeneity. Contributions to Statistics. Moda 4Advances in Model-Oriented Data Analysis 49-56. Springer, New York. · Zbl 0840.62079
[13]	Galtchouk, L. I. and Konev, V. I. (1997). On sequential estimation of parameters in continuous time stochastic regression. Statistics and Control of Stochastic Processes. The R. Lipster Festschrift 123-138. World Scientific, Singapore. · Zbl 0938.62083
[14]	Greenwood, P. and Shiryaev, A. N. (1992). Asymptotic minimaxity of a sequential estimator for a first order autoregressive model. Stochastics 38 49-65. · Zbl 0753.62049
[15]	Jacod, J. (1979). Calcul stochastique et problemes de martingales. Lecture Notes in Math. 714. Springer, New York. · Zbl 0414.60053 · doi:10.1007/BFb0064907
[16]	Jaukunas, A. and Khasminskii, R. Z. (1997). Estimation of parameters of linear homogeneous stochastic differential equations. Stochastic Processes Appl. 72 205-219. · Zbl 0933.62099 · doi:10.1016/S0304-4149(97)00083-5
[17]	Konev, V. V. (1985). Sequential Parameter Estimation of Stochastic Dynamical Systems. Tomsk Univ. Publishing House, Tomsk. (In Russian.) · Zbl 0582.93060
[18]	Konev, V. V. and Lai, T. L. (1995). Estimators with prescribed precision in stochastic regression models. Sequential Anal. 14 179-192. · Zbl 0838.62072 · doi:10.1080/07474949508836330
[19]	Konev, V. V. and Pergamenshchikov S. (1981). Sequential identification plans for dynamic systems. Automation Remote Control 42 917-924. · Zbl 0494.93040
[20]	Konev, V. V. and Pergamenshchikov, S. (1992). Sequential estimation of parameters in unstable stochastic system with guaranteed mean-square accuracy. Problems Inform. Trans. 28 35-48. · Zbl 0801.62069
[21]	Konev, V. V. and Pergamenshchikov,S. (1997). On guaranted estimation of the mean of an autoregressive process. Ann. Statist. 25 2127-2163. · Zbl 0887.62087 · doi:10.1214/aos/1069362391
[22]	Konev, V. V. and Vorobeichikov, S. E. (1980). On sequential identification of stochastic systems. Izvestia Acad. USSR Technical Cybern. N4 176-182. · Zbl 0465.93081
[23]	Lai, T. L., Robbins, H. and Wei, C. Z. (1979). Strong consistency of least squares estimates in multiple regression II. J. Multivariate Anal. 9 343-361. · Zbl 0416.62051 · doi:10.1016/0047-259X(79)90093-9
[24]	Lai, T. L. and Wei, C. Z. (1982). Least square estimates in stochastic regression regression models with application to identification and control of dynamic systems. Ann. Statist. 10 154- 166. · Zbl 0649.62060 · doi:10.1214/aos/1176345697
[25]	Lai, T. L. and Wei, C. Z. (1983). Asymptotic properties of general autoregressive models and strong consistency of least square estimates of their parameters. J. Multivariate Anal. 13 1-23. · Zbl 0509.62081 · doi:10.1016/0047-259X(83)90002-7
[26]	Lai, T. L. and Ziegmund, D. (1983). Fixed-accuracy estimation of an autoregressive parameter. Ann. Statist. 11 478-485. · Zbl 0519.62076 · doi:10.1214/aos/1176346154
[27]	Le Breton, A. (1977). Parameter estimation in a linear stochastic differential equation. Proceedings of the 7th Prague Conference 15-32. Chechoslovak Academy of Science, Prague. · Zbl 0413.60052
[28]	Le Breton, A. and Musiela, M. (1987). Strong consistency of LS-estimates in linear regression models driven by semimartingales. J. Multivariate Anal. 23 77-92. · Zbl 0627.62089 · doi:10.1016/0047-259X(87)90179-5
[29]	Lipster, R. Sh. and Shiryaev, A. N. (1977). Statistics of Random Processes. I, II. Springer, New York. · Zbl 0364.60004
[30]	Lipster, R. Sh. and Shiryaev, A. N. (1989). Theory of Martingales. Kluwer, Dordrecht.
[31]	Melnikov, A. V. (1996). Stochastic differential equations: nonsmoothness of coeffiicients, regression models and stochastic approximation. Russian Math. Surveys 51 43-136. · Zbl 0962.60041 · doi:10.1070/RM1996v051n05ABEH002994
[32]	Melnikov, A. V. and Novikov, A. A. (1988). Sequential inference with fixed accuracy for semimartigales. Theory Probab. Appl. 33 480-494. · Zbl 0693.62066 · doi:10.1137/1133071
[33]	Meyer, P. A. (1975). Un cours sur les integrales stochastiques. Seminaire de Probabilites X. Lecture Notes in Math. 511. Springer, New York. · Zbl 0374.60070
[34]	Novikov, A. A. (1984). Consistency of least squares estimates in regression models with martingale errors. Statistics and Control of Stochastic Processes (N. V. Krylov, R. S. Liptser and A. A. Novikov, eds.) 389-409. Springer, New York. · Zbl 0645.62094
[35]	Rao, C. R. (1968). Linear Statistical Inferences and Its Applications. Wiley, New York. · Zbl 0169.21302
[36]	Shiryaev, A. N. and Spokoiny, V. G. (1997). On sequential estimation of an autoregressive parameter. Stochastics Stochastic Reports 60 219-240. · Zbl 0877.62077
[37]	Sriram, T. N. (1988). Sequential estimation of the autoregressive parameter in a first-order autoregressive process. Sequential Anal. 7 53-74. · Zbl 0669.62074 · doi:10.1080/07474948808836142
[38]	Stricker, Ch. (1981). Les intervalles de constance de X X. Seminaire de Probabilites XVI. Lecture Notes in Math. 920 219-220. Springer, New York. · Zbl 0485.60055