# Article

Full entry | PDF   (2.0 MB)
Keywords:
nonparametric methods; isotonic regression; quantile; recursive methods; $LD-50$ value; 0-1 observations; Robbins-Monro procedure; examples; stochastic approximation
Summary:
The objective of this paper is to introduce some recursive methods that can be used for estimating an $LD-50$ value. These methods can be used more generally for the estimation of the $\gamma$-quantile of an unknown distribution provided we have 0-1 observations at our disposal. Standard methods based on the Robbins-Monro procedure are introduced together with different approaches of Wu or Mukerjee. Several examples are also mentioned in order to demonstrate the usefulness of the methods presented.
References:
[1] Barlow R. E., Bartholomew D. J., Bremner J. M., and Brunk H.D.: Statistical Inference Under Order Restrictions. Wiley, London, 1972.
[2] Blum J. R.: Multidimensional stochastic approximation procedures. AMS 25 (1954), 737-744. MR 0065092
[3] Charamza P.: Stochastic approximation on the lattice. Diploma work, Faculty of mathematics and physics, Charles University Prague, 1984. (In Czech.)
[4] Charamza P.: Software for stochastic approximation. Proceedings of the conference ROBUST 1990.
[5] Derman C.: Non-parametric up-and-down experimentation. AMS 28 (1957), 795-797. MR 0090956 | Zbl 0084.14801
[6] Dixon Mood: A method for obtaining and analyzing sensitivity data. JASA 43 (1948), 109-126. DOI 10.1080/01621459.1948.10483254
[7] Dupač V.: Quasiisotonic regression and stochastic approximation. Metrika 34 (1987), 117-123. DOI 10.1007/BF02613138 | MR 0882505
[8] Dupač V., Henkenrath U.: On integer stochastic approximation. Aplikace matematiky 29 (1984), 372-383. MR 0772272
[9] Fabián V.: On asymptotic normality in stochastic approximation. AMS 39 (1968), 1327-1332. MR 0231429
[10] Holst U.: Recursive estimation of quantiles. Contributions to probability and statistics in honor of Gunnar Blom, Univ. Lund, 1985, pp. 179-188. MR 0795057 | Zbl 0575.62074
[11] Lord P. M.: Tailored testing. An application of stochastic approximation. Journal of the American Statistical Association 66 (1971), 707-711. DOI 10.1080/01621459.1971.10482333 | Zbl 0257.62049
[12] Mukerjee H. G.: A stochastic approximation by observation on a discrete lattice using isotonic regression. AMS 9 (1981), 1020-1025. MR 0628757
[13] Nevelson M. B.: On the properties of the recursive estimates for a functional of an unknown distribution function. Limit Theorems of Probability Theory (P. Revezs, eds.), Amsterodam, 1975, pp. 227-252. MR 0395113
[14] Robbins H., Lai T. L.: Adaptive design and stochastic approximation. AS 7 (1979), 1196-1221. MR 0550144 | Zbl 0426.62059
[15] Robbins H., Lai T. L.: Consistency and asymptotic efficiency of slope estimates in stochastic approximation schemes. Z. Wahrsch. Verw. Gebiete 56 (1981), 329-360. DOI 10.1007/BF00536178 | MR 0621117 | Zbl 0472.62089
[16] Roth, Josífko, Malý, Trčka: Statistical methods in experimental medicine. Státní. zdravotnické nakladatelství, Praha, 1962, pp. 592. (In Czech.)
[17] Sacks J.: Asymptotic distribution of stochastic approximation procedures. AMS 29 (1958), 373-405. MR 0098427 | Zbl 0229.62010
[18] Wu C.F. J.: Efficient sequential designs with binary data. Journal of the American Statistical Association 80 (1985), 974-984. DOI 10.1080/01621459.1985.10478213 | MR 0819603 | Zbl 0588.62133
[19] Silvapulle M. J.: On the existence of maximum likelihood estimators for the binomial response problem. Journal of the Royal Statistical Society, Ser. B 43 (1981), 310-313. MR 0637943

Partner of