×

Selection of populations: An overview and some recent results. (English) Zbl 0712.62018

Summary: In practice the problem of comparing several treatments can often be formulated as which treatment can be considered best. Selection procedures have been designed specifically to answer questions like this one. In many situations selection procedures are more realistic than the usual testing procedures. In this paper an introduction and an overview of selection of populations will be given with special reference to Normal, Binomial, Poisson and Multinomial populations. In the field of agricultural applications these distributions play an important role.
Bechhofer’s and Gupta’s approach are presented side by side for a number of problems. After that the relatively new procedures of P. N. Somerville [see J. Stat. Comput. Simulation 22, 27-50 (1985; Zbl 0606.62022)] are discussed. Finally, some recent results are presented.

MSC:

62F07 Statistical ranking and selection procedures

Citations:

Zbl 0606.62022
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] The section, where the reference is quoted, is indicated between.
[2] Alam, Selection from Poisson processes. Ann. Inst. Statist. Math. 23 pp 411– (1971) · Zbl 0255.62031 · doi:10.1007/BF02479240
[3] Amer, Two new goals for selection based on proportions. Commun. Statist. Theor. Meth. A 9 pp 1461– (1980) · Zbl 0444.62033 · doi:10.1080/03610928008827974
[4] Anderson, Classification by multivariate analysis. Psychometrika 16 pp 31– (1951)
[5] and (1964). Tables for the studentized largest chi-square distribution and their application s. ARL 69–188, Aerospace Research Laboratories, Wright-Patterson Air Force Base, Dayton, Ohio. [4.2.1]
[6] Bahadur, Ann. Math. Statist. 21 pp 362– (1950)
[7] Bahadur, The problem of the greater mean. Ann. Math. Statist. 21 pp 469– (1950) · Zbl 0040.07501
[8] Correction 22 pp 310– (1951)
[9] Bechhofer, Ann. Math. Statist. 25 pp 16– (1954)
[10] Bechhofer, A multiplicative model for analyzing variances which are affected by several factors. J. Amer. Statist. Assoc. 55 pp 245– (1960) · Zbl 0201.52501
[11] (1977). Selection in factorial experiments. Proc. of the 1977 Winter Simulation Conference (E d. H. J. Highland, R. G. Sargent, and J. W. Schmidt) held at the National Bureau of Standards, Gaithersburg, Maryland, pp. 65–70. [7]
[12] Bechhofer, J. Amer. Statist. Assoc. 75 pp 751– (1980a)
[13] (1980b). Indifference-zone approach to selection and rankin g. Lecture Notes for the Seminar on ”Selection and Ranking Procedures”, November 17–21, 1980, Berlin. [4.3.1]
[14] (1985). Selection and Ranking Procedures–Some personal reminiscences, and thoughts about its past, present, and future. American Journal of Mathematical and Management Sciences, Vol. 5, NOS. 3 & 4, 201–234. [1], [2] · Zbl 0613.62030
[15] Bechhofer, A two-sample multiple decision procedure for ranking means of normal populations with a common unknown variance. Biometrika 41 pp 170– (1954) · Zbl 0056.13002
[16] Bechhofer, A single-sample multiple decision procedure for selecting the multinomial event which has the highest probability. Ann. Math. Statist. 30 pp 102– (1959) · Zbl 0218.62064
[17] and (1968). Sequential Identification and Ranking Procedures. The University of Chicago Press, Chicago. [1], [4. 1. 1], [4. 2. 1], [7].
[18] and (1982a). Closed adaptive sequential procedures for selecting the best of k Bernoulli populations. Proc. Third Purdue Symp. Stat. Dec. Theory and Related Topics (Ed s. S. S. Gupta and R. L. Berger) I, pp. 61–108. [7]
[19] Bechhofer, Statist. 1 pp 315– (1982b)
[20] Bechhofer, Closed sequential procedures for selecting the multinomial events which have the largest probabilities. Commun. Statist. Theor. Meth. A 13 pp 2997– (1984) · Zbl 0571.62069
[21] and (1977). Selecting the largest interaction in a two-factor experiment. Statistical Decision Theory and related Topics-II (Eds. and ), Academic Press, New York, pp. 1–18. [7] · doi:10.1016/B978-0-12-307560-4.50005-3
[22] Bechhofer, A single-sample multiple-decision procedure for ranking variances of normal populations. Ann. Math. Statist. 25 pp 273– (1954) · Zbl 0055.13003
[23] Bechhofer, J. Amer. Statist. Assoc. 73 pp 385– (1978)
[24] Berger, Minimax subset selection rules with applications to unequal variance (unequal sample size) problems. Scand. J. Statist. 7 pp 21– (1980) · Zbl 0451.62023
[25] Birnbaum, Effect of linear truncation on a multinormal population. Ann. Math. Statist. 21 pp 272– (1950) · Zbl 0038.09201
[26] Birnbaum, On optimum selections from multinormal populations. Ann. Math. Statist. 21 pp 443– (1950) · Zbl 0039.14201
[27] and (1978). Statistics for Experimenters, John Wiley & Sons, New York. [1]
[28] Brostroem, On sequentially rejective subset selection procedures. Commun. Statist. Theor. Meth. A 10 pp 203– (1981)
[29] and (1980). Nonparametric Sequential Selection Procedures. Birkhuser, Boston. [1] · Zbl 0462.62023 · doi:10.1007/978-1-4684-0538-5_1
[30] and (1987). Tables for selecting the best populatio n. Queensland Biometrical Bulletin 2, Queensland Department of Primary Industries, Brisbane, QLD 4001, Australia. [1]
[31] and (1966). An inverse-sampling procedure for selecting the most probable event in a multinomial distribution. Multivariate Analysis, Ed. Academic Press, New York, pp. 423–455. [6. 1]
[32] Carroll, Commun. Statist. 4 pp 987– (1975)
[33] Chen, On selecting a subset which contains all populations better than a control. Commun. Statist. Theor. Meth. A 9 pp 851– (1980) · Zbl 0437.62031
[34] and (1976). Subset selection Procedures for normal means under unequal sample sizes. Sankhya Ser. B,38, 249–255. [4. 1. 2] · Zbl 0409.62021
[35] Chew, Statistical hypothesis testing: an academic exercise in futility. Proc. Fla. State Hort. Soc. 90 pp 214– (1977)
[36] (1951). Improvement by means of selection. Proceedings of the Second Berkeley Symposiumon Math. Statist. and Probability, University of California Press, Berkeley, pp. 449–470. [1]
[37] and (1957). Experimental Designs, second Edition. John Wiley & Sons, Inc., New York. [1]
[38] (1970). The analysis of binary data. Methuen, London. [7]
[39] Curnow, The consequences of errors of measurement for selection from certain non-normal distributions. Bull. Internat. Stat. Inst. 37 pp 291– (1960) · Zbl 0092.37001
[40] Dalal, Most economical robust selection procedures for location parameters. Ann. Statist. 7 pp 1321– (1979) · Zbl 0435.62026
[41] Davies, The design of screening tests in the pharmaceutical industry. Bull. Internat. Stat. Inst. 36 pp 226– (1958) · Zbl 0088.11904
[42] Deely, On the properties of subset selection procedures. Sankhya Ser. A 30 pp 37– (1968) · Zbl 0227.62013
[43] Desu, A selection problem. Ann. Math. Statist. 41 pp 1596– (1970) · Zbl 0226.62057 · doi:10.1214/aoms/1177696803
[44] (1969). A general selection Procedure relative to t best populations. Ph. D. Thesis, Dep t. of Statist., Purdue Univ., West Lafayette, Indiana. [7]
[45] Deverman, Also abstract in Ann. Math. Statist. 40 pp 7– (1969)
[46] , and (1987). Selection from Normal populations with unequal variances and its robustnes s. Technical Note 87–04, Dep t. of Mathematics, Statistics Division, Agricultural University, Wageningen. [9]
[47] Driessen, Submitted for publication to Biom. J. 9 (1988)
[48] (1976). Introduction to Statistics and Probability. Holt, Rinehart and Winston, New York. [1] · Zbl 0358.62001
[49] Dudewicz, Technometrics 21 pp 582– (1979)
[50] Dudewicz, Ranking (ordering) and selection: An overview of how to select the best. Technometrics 22 pp 113– (1980) · Zbl 0438.62023
[51] Dudewicz, Review of ”Multiple Decision Procedures: Theory and Methodology of Selecting and Ranking Populations” by Gupta and Panchapakesan. Technometrics 24 pp 162– (1982)
[52] Dudewicz, Sankhya Ser. B 37 pp 28– (1975)
[53] and (1982). The Complete Categorized Guide to Statistical Selection and Ranking Procedures. Series in Mathematical and Management Sciences, Vol. 6, American Sciences Press, Inc., Columbus, Ohio. [1] · Zbl 0509.62021
[54] and (1980). Ranking and selection in designed experiments: complete factorial experiments. Tech n. Report 210, Ohio, State University. [7]
[55] Dunnett, J. Roy. Stat. Soc. Ser. B 22 pp 1– (1960)
[56] Dunnett, Biometrika 41 pp 153– (1954) · Zbl 0056.36703 · doi:10.1093/biomet/41.1-2.153
[57] Federer, Biometrics 19 pp 553– (1963)
[58] Finney, The consequences of selection for a variate subject to errors of measurement. Rev. Inst. Internat. Stat. 24 pp 1– (1956) · Zbl 0073.35604
[59] Finney, Statistical problems in plant selection. Bull. Internat. Stat. Inst. 36 pp 242– (1958) · Zbl 0087.15604
[60] and (1977). Selecting and Ordering Populations: A New Statistical Methodology. John Wiley & Sons, Inc., New York. [1], [2], [3], [4. 1. 1], [4. 1. 2], [4. 3. 1], [4. 3. 2], [4. 4. 1], [5. 1. 1]. [5. 1. 2], [5. 2. 1], [6], [6. 1], [7] · Zbl 0464.62022
[61] Gibbons, An introduction to ranking and selection. The American Statistician 33 pp 185– (1979)
[62] Girshick, Contributions to the theory of sequential analysis: I. Ann. Math. Statist. 17 pp 123– (1946) · Zbl 0063.01636
[63] (1972). A note on the non-existence of subset selection Procedure for Poisson populations. Mimeo. Ser. No. 303. Dep t. of Statist., Purdue Univ., West Lafayette, Indiana. [4.2.2]
[64] (1956). On a decision rule for a problem in ranking means. P h. D. Dissertation (and Mimeograph Series No. 150), Institute of Statistics, University of North Carolina, Chapel Hill. [1], [3], [4. 3. 2], [5. 1. 2]
[65] Gupta, Ann. Math. Statist. 34 pp 792– (1963a)
[66] Gupta, On a selection and ranking procedure for gamma populations. Ann. Inst. Statist. Math. 14 pp 199– (1963b) · Zbl 0128.13201 · doi:10.1007/BF02868642
[67] Gupta, On some multiple decision (selection and ranking) rules. Technometrics 7 pp 225– (1965) · Zbl 0147.17902
[68] Gupta, Commun. Statist. Theor. Meth. A6 pp 993– (1977a)
[69] Gupta, Statist. Theor. Meth. A 6 (1977b)
[70] (1980). Selection and ranking procedures, lectures on subset selection approach and related topic s. Lecture Notes for the Seminar on ”Selection and Ranking Procedures”, November 17–21, 1980, Berlin. [4.3.2]
[71] and (1975). On subset selection Procedures for Poisson populations and some applications to the multinomial selection problems. Applied Statistics (E d. R. P. Gupta), North-Holland, Amsterdam, pp. 97–109. [4.2.2]
[72] Gupta, Sankhya Ser. B 38 pp 112– (1976a)
[73] and (1976b). On some methods for constructing optimal subset selection Procedures. Mimeo-Ser. No. 470, Dep t. of Statist., Purdue Univ., West Lafayette, Indiana. [4.4.2]
[74] Gupta, A note on optimal subset selection procedures. Ann. Statist. 8 pp 1164– (1980a) · Zbl 0459.62019
[75] Gupta, An essentially complete class of multiple decision procedures. J. Statist. Planning and Inference 4 pp 115– (1980b) · Zbl 0443.62016 · doi:10.1016/0378-3758(80)90001-4
[76] , and (1976). On ranking and selection Procedures and tests of homogeneity for binomial populations. Essays in Probability and Statistics (Eds. S. Ikeda et al.), Shinko Tsusho Co. Ltd., Tokyo, Japan, pp. 501–533. [4. 1. 2]
[77] Gupta, Sankhya Ser. A 36 pp 389– (1974)
[78] Gupta, Sankhya Ser. B 43 pp 245– (1981)
[79] Gupta, Sankhya Ser. B 29 pp 1– (1967)
[80] Gupta, On the order statistics from equally correlated normal random variables. Biometrika 60 pp 403– (1973) · Zbl 0263.62031
[81] Gupta, On a class of subset selection procedures. Ann. Math. Statist. 43 pp 814– (1972a) · Zbl 0251.62016
[82] Gupta, On multiple decision (subset selection) procedures. J. of Math. and Physical Sciences 6 pp 1– (1972b) · Zbl 0336.62019
[83] and (1979). Multiple Decision Procedures Theory and Methodology of Selecting and Ranking Populations. John Wiley and Sons, New York–London. [1], [2], [3], [4. 1. 2], [5. 1. 2], [5. 2. 2], [6] · Zbl 0516.62030
[84] and (1985). Subset Selection Procedures: Review and assessment. American Journal of Mathematical and Management Sciences, Vol. 5, NOS. 3 & 4, 235–312. [1], [3] · Zbl 0633.62024
[85] and (1973). On selection and ranking procedures–a restricted subset selection rule. Proceedings of the 39th Session of the ISI, vol. 45, book 1, pp. 409–417. [7]
[86] and (1979). On selection rules for treatments versus control problems. Contributed paper on ISI conference, Manilla, 229–232. [7]
[87] Gupta, On rules based on sample medians for selecting of the largest location parameter. Commun. Statist. Theor. Meth. A 9 pp 1277– (1980) · Zbl 0443.62015 · doi:10.1080/03610928008827958
[88] Gupta, On a statistic which arises in selection and ranking problems. Ann. Math. Statist. 28 pp 957– (1957) · Zbl 0080.35504
[89] and (1960). Selecting a subset containing the best of several binomial populations. Contributions to Probability and Statistics (Eds. et al.), Stanford University Press, Stanford, California, pp. 224–248. [4. 1. 2]
[90] Gupta, On selecting a subset containing the population with the smallest variance. Biometrika 49 pp 495– (1962a) · Zbl 0114.35003
[91] Gupta, Biometrika 49 pp 509– (1962b) · Zbl 0124.11207 · doi:10.1093/biomet/49.3-4.509
[92] and (1976). Subset selection Procedures for the means of normal populations with unequal variances: unequal sample sizes cas e. Mimeo, Ser. No. 473, Dep t. of Statist., Purdue Univ., West Lafayette, Indiana. [5.1.2]
[93] and (1975). A survey of adaptive sampling for clinical trials. In Perspectives in Biometrics, Volume 1, Ed. Academic Press, New York – London, pp. 29–61. [1] · doi:10.1016/B978-0-12-237301-5.50007-5
[94] Huang, Commun. Statist. Theor. Meth. A 12 pp 341– (1983)
[95] Jennison, Equal probability of correct selection for Bernoulli selection procedures. Commun. Statist. Theor. Meth. A 12 pp 2887– (1983) · Zbl 0537.62019
[96] Kesten, A property of the multinomial distribution. Ann. Math. Statist. 30 pp 120– (1959) · Zbl 0218.62017 · doi:10.1214/aoms/1177706363
[97] (1975). Statistical Techniques in Simulation, part II, M. Dekker, New York. [1]
[98] and (1966). Some Procedures for selection of multivariate normal populations better than a control. Multivariate Analysis (Ed. P. R. Krishnaiah), Academic Press, New York, pp. 477–490. [7]
[99] Van der Laan, Tom pp 203– (1987a)
[100] Van der Laan, To appear in Statistica Neerlandica 43 (1987b)
[101] and (1988). The robustness of the probability of correct selection against various distributional deviation s. Technical Report, Dep t. of Mathematics, Statistics Division, Agricultural University, Wageninge n. Prepublication intended as publication in Commun. Statist. [9]
[102] Lorenzen, Selecting logistic populations using the sample medians. Commun. Statist. Theor. Meth. A 10 pp 101– (1981) · Zbl 0468.62024 · doi:10.1080/03610928108828024
[103] (1977). Selection in factorial experiment s. Project Report, Statistics 828, Ohio State University. [7]
[104] Mahamunulu, Some fixed-sample ranking and selection problems. Ann. Math. Statist. 38 pp 1079– (1967) · Zbl 0158.36702
[105] McDonald, Subset selection procedures based on sample ranges. Commun. Statist. Theor. Meth. A 6 pp 1055– (1977) · Zbl 0375.62028 · doi:10.1080/03610927708827551
[106] (1971). Computer Simulation Experiments with Models of Economic Systems. John Wiley & Sons, Inc., New York. [1]
[107] (1971). On a subset selection procedure for the most probable event in a multinomial distribution. Statistical Decision Theory and Related Topics, (Eds. and ), Academic Press, New York, pp. 275–298. [6. 2] · doi:10.1016/B978-0-12-307550-5.50019-8
[108] Panchapakesan, Inst. Math. Statist. 2 pp 112– (1973)
[109] Panchapakesan, Commun. Statist. Theor. Meth. A 6 pp 1081– (1977)
[110] Parnes, Sankhya Ser. A 48 pp 86– (1986)
[111] Paulson, A multiple decision procedure to certain problems in analysis of variance. Ann. Math. Statist. 20 pp 239– (1949) · Zbl 0032.04403
[112] Rasch, Selection problems in Balanced Block Designs. Biometrical Journal 20 pp 275– (1978) · Zbl 0379.62066
[113] (1984). Einfhrung in die mathematische Statistik II. Varianzanalyse, Regressionsanalyse und weitere Anwendungen. VEB Deutscher Verlag der Wissenschaften, Berlin. [1] · Zbl 0534.62047
[114] (1963). Ranking and selection problems of normal populations using the absolute values of their means: fixed sample size cas e. Technical Report No. 31, Dep t. of Statist., Uni v. of Minnesota, Minneapolis, Minnesota. [4. 3. 2]
[115] Rizvi, Some selection problems involving folded normal distribution. Technometrics 13 pp 355– (1971) · Zbl 0218.62059
[116] Rizvi, M. H. (Ed.) (1985). Modern Statistical Selection, Part I–Robert E. Bechhofer, Shanti S. Gupta, and Milton Sobel, Proceedings of the Conference Statistical Ranking and Selection – Three Decades of Development, Univ. of California at Santa Barbara, December 1984. American Journal of Mathematical and Management Sciences, Vol. 5, NOS. 3 & 4. [1]
[117] Rizvi, M. H. (Ed.) (1986). Modern Statistical Selection, Part II – Robert E. Bechhofer, Shanti S. Gupta, and Milton Sobel, Proceedings of the Conference Statistical Ranking and Selection – Three Decades of Development, Univ. of California at Santa Barbara, December 1984. American Journal of Mathematical and Management Sciences, Vol. 6, NOS. 1 & 2. [1]
[118] Santner, Ann. Statist 3 pp 334– (1975)
[119] and (1984). Design of Experiments: Ranking and Selection. Marcel Dekker, Basel and New York. [1]
[120] Sieben, Selectie bij de plantenveredeling op grond van proefveldresultaten. Statistica Neerlandica 8 pp 179– (1954) · doi:10.1111/j.1467-9574.1954.tb00450.x
[121] (1963). Simple-sample ranking problems with Poisson populations. Techn. Report No. 19, Dept. of Statist., Univ. of Minnesota, Minneapolis. [4.2.1]
[122] (1969). Selecting a subset containing at least one of the t best populations. Multivariate Analysis-II (Ed. ), Academic Press, New York, pp. 515–540. [7]
[123] Sobel, American Journal of Mathematical and Management Sciences 5 pp 313– (1985) · Zbl 0658.62034 · doi:10.1080/01966324.1985.10737172
[124] Sobel, Selecting the best one of several binomial populations. Bell System. Tech. J. 36 pp 537– (1957) · doi:10.1002/j.1538-7305.1957.tb02411.x
[125] Somerville, A multiple range subset selection procedur e. J. Statist. Comput. Simul. 19 pp 215– (1984) · doi:10.1080/00949658408810733
[126] Somerville, A new subset selection method for Normal population s. J. Statist. Comput. Simul 22 pp 27– (1985a) · Zbl 0606.62022 · doi:10.1080/00949658508810831
[127] (1985b). Subset selection, not multiple comparisons. Proceedings of the Tenth Annual SAS Users Group Intern. Conf., Reno, Nevada, pp. 1113–1115. [8]
[128] Szanti, Alkamaz. Mat. Lapok. 2 pp 27– (1976)
[129] Tamhane, Sankhya Ser. B 39 pp 344– (1977)
[130] Tamhane, Selecting the better Bernoulli treatment using a matched samples design. J. R. Statist. Soc. B 42 pp 26– (1980) · Zbl 0425.62015
[131] Tamhane, J. Amer. Statist. Assoc 80 pp 455– (1985)
[132] Tamhane, A two-stage minimax procedure with screening for selecting the largest normal mean. Commun. Statist. Theor. Meth. A 6 pp 1003– (1977)
[133] Tamhane, A two-stage minimax procedure with screening for selecting the largest normal mean (II): An improved PCS lower bound and associated tables. Commun. Statist. Theor. Meth. A 8 pp 337– (1979) · Zbl 0442.62028
[134] (1947). Sequential Analysis. John Wiley, New York. [1] · Zbl 0029.15805
[135] Wetherill, Appl. Statist. 23 pp 253– (1974)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.