Gradus

VOL 2, NO 2 (2015): AUTUMN (NOVEMBER)

 

Eloszláscsaládokhoz való illeszkedés vizsgálata

Goodness of fit to family of distribution


Osztényiné Krauczi Éva

Abstract

Eloszláscsaládokhoz való illeszkedés vizsgálata a matematikai statisz- tikának a hipotézisvizsgálathoz tartozó területe. Ennek az összefoglaló cikknek az a célja, hogy áttekintést adjon az ezen a területen elért leg- elso? eredményekro?l. Felidézzük az elso? teszteket, amelyekkel rögzített eloszláshoz való illeszkedést lehet elleno?rizni, valamint, hogy hogyan találták meg ezeknek a tesztstatisztikáknak a határeloszlásait. Majd az elso? összetett illeszkedésvizsgálati módszereket és határeloszlásukat elevenítjük fel. Ezen eljárások két nagy osztályát tárgyaljuk részlete- sen, az egyik a minta eloszlásának és az eloszláscsalád eloszlásainak távolságán alapuló tesztek, a másik a regresszió-, illetve korrelációtesz- tek.

Goodness of fit to family of distribution belongs to hypothesis tests of mathematical statistics. The goal of this paper is to give a summary of the first results of this area. For the overview we recall the first tests which are suitable for goodness of fit to a fixed distribution paying spe- cial attention to the development of the asymptotic theory of goodness of fit tests. The goodness of fit to family of distributions and their asymp- totic theories are considered, focusing on two classes of this procedure: tests of fit based on the empirical distribution function (EDF), and the regression and correlation tests of fit.


Keywords

Kulcsszavak: illeszkedésvizsgálat, határeloszlás-tétel,

Keywords: goddness-of-fit, asymptotic distribution theorem,


References

[1] M. M. Ali. Stochastic ordering and kurtosis measure. Journal of the American Statistical Association,69:543–545, 1974.
[2] T. W. Anderson and D. A. Darling. Asymptotic theory of certain „goodness of fit” criteria based on sto-chastic processes. Ann. Math. Statistics, 23:193–212, 1952.
[3] P. Billingsley. Convergence of probability measures. New York-London-Sydney-Toronto: John Wiley andSons, Inc. XII, 1968.
[4] M. Burke, M. Csörgo?, S. Csörgo?, and P. Révész. Approximations of the empirical process when parame-ters are estimated. Ann. Probab., 7(5):790–810, 1979.
[5] H. Chernoff and E. Lehmann. The use of maximum likelihood estimates in ?2 tests for goodness of fit.Ann. Math. Stat., 25:579–586, 1954.
[6] W. Cochran. The ?2 test of goodness of fit. Annals of Mathematical Statistics, 23:315–345, 1952.
[7] H. Cramér. On the composition of elementary errors. I. Mathematical deductions. II. Statistical applica-tions. Skand. Aktuarietidskr., 11:13–74, 141–180, 1928.
[8] S. Csörgo?. Limit behaviour of the empirical characteristic function. Ann. Probab., 9:130–144, 1981.
[9] S. Csörgo?. The empirical characteristic process when parameters are estimated. Contributions to proba-bility, Collect. pap. dedic. E. Lukacs, 215-230 , 1981.
[10] S. Csörgo?. Testing for normality in arbitrary dimension. Annals of Statistics, 14:708–723, 1986.
[11] S. Csörgo?. Consistency of some tests for multivariate normality. Metrika, 36:107–116, 1989.
[12] R. B. D’Agostino. An omnibus test of normality for moderate and large sample sizes. Biometrika, 58:341–348, 1971.
[13] D. Darling. The Cramér-Smirnov test in the parametric case. Ann. Math. Stat., 26:1–20, 1955.
[14] F. David and N. Johnson. The probability integral transformation when parameters are estimated from thesample. Biometrika, 35:182–190, 1948.
[15] H. David, H. Hartley, and E. Pearson. The distribution of the ratio, in a single normal sample, of range tostandard deviation. Biometrika, 41:482–493, 1954.
[16] T. de Wet and J. Venter. Asymptotic distributions of certain test criteria of normality. S. Afr. Stat. J.,6:135–149, 1972.
[17] E. del Barrio, J. A. Cuesta-Albertos, and C. Matrán. Contributions of empirical and quantile processes tothe asymptotic theory of goodness-of-fit tests. Test, 9(1):1–96, 2000. With discussion.
[18] M. D. Donsker. An invariance principle for certain probability limit theorems. Mem. Am. Math. Soc., 6:12,1951.
[19] J. L. Doob. Heuristic approach to the Kolmogorov-Smirnov theorems. Ann. Math. Stat., 20:393–403,1949.
[20] J. Durbin. Weak convergence of the sample distribution function when parameters are estimated. Ann.Stat., 1:279–290, 1973.
[21] T. Epps and L. B. Pulley. A test for normality based on the empirical characteristic function. Biometrika,70:723–726, 1983.
[22] W. Feller. On the Kolmogorov-Smirnov limit theorems for empirical distributions. Ann. Math. Stat., 19:177–189, 1948.
[23] A. Feuerverger and R. A. Mureika. The empirical characteristic function and its applications. Ann. Stat.,5:88–97, 1977.
[24] R. A. Fisher. The moments of the distribution for normal samples of measures of departure from normality.Proc. R. Soc. Lond., Ser. A, 130:16–28, 1930.
[25] R. Geary. Testing for normality. Biometrika, 34:209–242, 1947.
[26] E. Gumbel. On the reliability of the classical chi-square test. Ann. Math. Stat., 14:253–263, 1943.
[27] P. Hall and A. H. Welsh. A test for normality based on the empirical characteristic function. Biometrika,70:485–489, 1983.
[28] M. Kac, J. Kiefer, and J. Wolfowitz. On tests of normality and other tests of goodness of fit based ondistance methods. Ann. Math. Stat., 26:189–211, 1955.
[29] A. Kolmogorov. Sulla determinazione empirica di una legge di distribuzione. Giornale del Istituto Italianodegli Attuari, 4:83–91, 1933.
[30] É. Krauczi. A study of the quantile correlation test of normality. Test, 18(1):156–165, 2009.
[31] V. LaRiccia and D. M. Mason. Cramér-von Mises statistics based on the sample quantile function andestimated parameters. J. Multivariate Anal., 18:93–106, 1986.
[32] J. Leslie, M. Stephens, and S. Fotopoulos. Asymptotic distribution of the Shapiro-Wilk W for testing fornormality. Ann. Stat., 14:1497–1506, 1986.
[33] H. W. Lilliefors. On the kolmogorov-smirnov test for normality with mean and variance unknown. Journalof the American Statistical Association, 62:399–402, 1967.
[34] H. Mann and A. Wald. On the choice of the number of class intervals in the application of the chi squaretest. Ann. Math. Stat., 13:306–317, 1942.
[35] E. Pearson, R. D’Agostino, and K. Bowman. Tests for departure from normality: Comparison of powers.Biometrika, 64:231–246, 1977.
[36] E. S. Pearson. A further development of tests for normality. Biometrika, 22:239–249, 1930.
[37] D. Pollard. The minimum distance method of testing. Metrika, 27:43–70, 1980.
[38] M. W. Shapiro, S.S. and H. Chen. An approximate analysis of variance test for normality. Journal of theAmerican Statistical Association, 63:1343–72, 1968.
[39] S. Shapiro and R. Francia. An approximate analysis of variance test for normality. Journal of the AmericanStatistical Association, 67:215–216, 1972.
[40] S. Shapiro and M. Wilk. An analysis of variance test for normality (complete samples). Biometrika,52:591–611, 1965.
[41] N. Smirnov. Sur la distribution de ?2 (Critérium de M.R. von Mises). Comptes Rendus de l’Académie desSciences Paris, 202:449–452, 1936.
[42] N. Smirnov. Sur la distribution de ?2 (Critérium de M.R. von Mises). Matematicheskij Sbornik, 2:973–993,1937.
[43] N. Smirnov. Sur les écarts de la courbe de distribution empirique. Matematicheskij Sbornik, 6:3–26, 1939.
[44] N. Smirnov. Approximate laws of distribution of random variables from empirical data. Uspekhi Matema-ticheskikh Nauk, 10:179–206, 1941.
[45] M. A. Stephens. Edf statistics for goodness of fit and some comparisons. Journal of the AmericanStatistical Association, 69:730–737, 1974.
[46] S. Sukhatme. Fredholm determinant of a positive definite kernel of a special type and its application. Ann.Math. Stat., 43:1914–1926, 1972.
[47] S. Verrill and R. Johnson. The asymptotic equivalence of some modified shapiro-wilk statistics - completeand censored sample cases. Annals of Statstics, 15:413–419, 1987.
[48] R. von Mises. Wahrscheinlichkeitsrechnung. Wein, Leipzig, 1931.
[49] G. Watson. The ?2 goodness-of-fit test for normal distributions. Biometrika, 44:336–348, 1957.
[50] G. Watson. On chi-square goodness-of-fit tests for continuous distributions. J. R. Stat. Soc., Ser. B,20:44–72, 1958.
[51] S. Weisberg and B. C. An approximate analysis of variance test for non-normality suitable for machinecalculation. Technometrics, 17:133–134, 1975.
[52] P. Williams. Note on the sampling distribution of where the population is normal. Biometrika, 27:269–271, 1935.



Copyright (c) 2019 Gradus