This article compares the size of selected subsets using nonparametric subset selection rules with two different scoring rules for the observations. The scoring rules are based on the expected values of order statistics of the uniform distribution (yielding rank values) and of the normal distribution (yielding normal score values). The comparison is made using state motor vehicle traffic fatality rates, published in a 2016 article, with fifty-one states (including DC as a state) and over a nineteen-year period (1994 through 2012). The earlier study considered four block design selection rules—two for choosing a subset to contain the “best” population (i.e., state with lowest mean fatality rate) and two for the “worst” population (i.e., highest mean rate) with a probability of correct selection chosen to be 0.90. Two selection rules based on normal scores resulted in selected subset sizes substantially smaller than corresponding rules based on ranks (7 vs. 16 and 3 vs. 12). For two other selection rules, the subsets chosen were very close in size (within one). A comparison is also made using state homicide rates, published in a 2022 article, with fifty states and covering eight years. The results are qualitatively the same as those obtained with the motor vehicle traffic fatality rates.
References
[1]
Conover, W.J. (1999) Practical Nonparametric Statistic. 3rd Edition, John Wiley & Sons, Inc., New York.
McDonald, G.C. (1972) Some Multiple Comparison Selection Procedures Based on Ranks.Sankhya: The Indian Journal of Statistics Series A, 34, 53-64.
[4]
McDonald, G.C. (1973) The Distribution of Some Rank Statistics with Applications in Block Design Selection Problems. Sankhya: The Indian Journal of Statistics SeriesA, 35, 187-204.
[5]
McDonald, G.C. (1979) Nonparametric Selection Procedures Applied to State Traffic Fatality Rates. Technometrics, 21, 515-523. https://doi.org/10.1080/00401706.1979.10489822
[6]
Lorenzen, T.J. and McDonald, G.C. (1984) A Nonparametric Analysis of Urban, Rural, and Interstate Traffic Fatality Rates. In: Santner, T.J. and Tamhane, A.C., Eds., Design of Experiments-Ranking and Selection, Marcel Dekker, New York, 143-178.
[7]
Green, J., McDonald, G.C. and Rao, N. (2006) Using Selection Procedures to Analyze State Traffic Fatality Rates. American Journal of Mathematical and Management Sciences, 26, 387-416. https://doi.org/10.1080/01966324.2006.10737680
[8]
Green, J. and McDonald, G.C. (2009) Nonparametric Subset Selection Procedures: Applications and Properties. American Journal of Mathematical and ManagementSciences, 29, 413-436. https://doi.org/10.1080/01966324.2009.10737766
[9]
McDonald, G.C. (2016) Applications of Subset Selection Procedures and Bayesian Ranking Methods in Analysis of Traffic Fatality Data. WIREs Computational Statistics, 8, 222-237. https://doi.org/10.1002/wics.1385
[10]
Wang, A.Q. and McDonald, G.C. (2022) Analysis of State Homicide Rates Using Statistical Ranking and Selection Procedures.Applied Mathematics, 13, 585-601. https://doi.org/10.4236/am.2022.137037
[11]
Gupta, S.S. and Panchapakesan, S. (1979) Multiple Decision Procedures. John Wiley & Sons, Inc. Republished in the Classics in Applied Mathematics Series, No. 44 (2002), Society for Industrial and Applied Mathematics, Philadelphia. https://epubs.siam.org/doi/pdf/10.1137/1.9780898719161.fm
[12]
McDonald, G.C. (2021) Computing Probabilities for Rank Statistics Used with Block Design Nonparametric Subset Selection Rules. American Journal of Mathematical and Management Sciences, 41, 38-50. https://doi.org/10.1080/01966324.2021.1910885
[13]
Harter, H.L. (1961) Expected Values of Normal Order Statistics. Biometrika, 48, 151-165. https://doi.org/10.1093/biomet/48.1-2.151
[14]
Birnbaum, A. and Dudman, J. (1963) Logistic Order Statistics. Annals of Mathematical Statistics, 34, 658-663. https://doi.org/10.1214/aoms/1177704178
[15]
Tukey, J.W. (1949) One Degree-of-Freedom for Non-Additivity. Biometrics, 5, 232-242. https://doi.org/10.2307/3001938