Goodness-of-fit Testing for Left-truncated Two-parameter Weibull Distributions with Known Truncation Point
DOI:
https://doi.org/10.17713/ajs.v45i3.106Abstract
The left-truncated Weibull distribution is used in life-time analysis, it has many applications ranging from financial market analysis and insurance claims to the earthquake inter-arrival times.We present a comprehensive analysis of the left-truncated Weibull distribution when the shape, scale or both parameters are unknown and they are determined from the data using the maximum likelihood estimator. We demonstrate that if both the Weibull parameters are unknown then there are sets of sample configurations, with measure greater than zero, for which the maximum likelihood equations do not possess non trivial solutions.
The modified critical values of the goodness-of-fit test from the Kolmogorov-Smirnov test statistic when the parameters are unknown are obtained from a quantile analysis. We find that the critical values depend on sample size and truncation level, but not on the actual Weibull parameters. Confirming this behavior, we present a complementary analysis using the Brownian bridge approach as an asymptotic limit of the Kolmogorov-Smirnov statistics and find that both approaches are in good agreement. A power testing is performed for our Kolmogorov-Smirnov goodness-of-fit test and the issues related to the left-truncated data are discussed. We conclude the paper by showing the importance of left-truncated Weibull
distribution hypothesis testing on the duration times of failed marriages in the US, worldwide terrorist attacks, waiting times between stock market orders, and time intervals of radioactive decay.
References
bibitem[{Agostino and Stephens(1986)}]{Dagostino1986}
Agostino R, Stephens MA (1986).
newblock emph{Goodness-of-fit techniques}.
newblock M. Dekker, New York.
newblock ISBN 0-8247-7487-6.
bibitem[{Aho emph{et~al.}(1985)Aho, Bain, and Engelhardt}]{Aho1985}
Aho M, Bain LJ, Engelhardt M (1985).
newblock enquote{Goodness-of-fit tests for the weibull distribution with
unknown parameters and heavy censoring.}
newblock emph{Journal of Statistical Computation and Simulation.},
textbf{21(3-4)}, 213--225.
bibitem[{Anderson and Stephens(1997)}]{AndersonStephens1997}
Anderson T, Stephens M (1997).
newblock enquote{The continuous and discrete Brownian bridges:
Representations and applications.}
newblock emph{Linear Algebra and its Applications}, textbf{264 (6)},
--171.
bibitem[{Balakrishnan and Cohen(1991)}]{Balakrishnan1991}
Balakrishnan N, Cohen A (1991).
newblock emph{Order Statistics and Inference: Estimation Methods}.
newblock Academic Press, Boston.
bibitem[{Balakrishnan and Mitra(2012)}]{Mitra2012}
Balakrishnan N, Mitra D (2012).
newblock enquote{Left truncated and right censored Weibull data and
likelihood inference with an illustration.}
newblock emph{Computational Statistics & Data Analysis}, textbf{56}, 4011
-- 4025.
bibitem[{Berkson(1975)}]{Berkson1975}
Berkson J (1975).
newblock enquote{Do Radioactive Decay Events Follow a Random
Poisson-Exponential?}
newblock emph{International Journal of Applied Radiation and Isotopes,},
textbf{26(9)}, 543--549.
bibitem[{Birnbaum(1952)}]{Birnbaum1952}
Birnbaum FJ (1952).
newblock enquote{Numerical Tabulation of the Distribution of Kolmogorov's
Statistic for Finite Sample Size.}
newblock emph{Journal of the American Statistical Association},
textbf{47}(259), 425--441.
bibitem[{Chandra emph{et~al.}(1981)Chandra, Singapurwalla, and
Stephens}]{Chandra1981}
Chandra M, Singapurwalla N, Stephens M (1981).
newblock enquote{Kolmogorov statistics for tests of fit for the extreme value
and Weibull distributions.}
newblock emph{Journal of the American Statistical Association}, textbf{76
(375)}, 729--731.
bibitem[{Clauset and Woodard(2013)}]{Clauset2013}
Clauset A, Woodard R (2013).
newblock enquote{Estimating the historical and future probabilities of large
terrorist events.}
newblock emph{Annals of Applied Statistics}, textbf{7 (4)}, 1838--1865.
bibitem[{Cohen(1965)}]{Cohen1965}
Cohen CA (1965).
newblock enquote{Maximum likelihood estimation in the Weibull distribution
based on complete and on censored samples.}
newblock emph{Technometrics}, textbf{7 (4)}, 579--588.
bibitem[{David and Johnson(1948)}]{DavidJohnson1948}
David F, Johnson N (1948).
newblock enquote{The Probability Integral Transformation When Parameters are
Estimated from the Sample.}
newblock emph{Biometrika}, textbf{35 (1/2)}, 182--190.
bibitem[{Durbin(1973)}]{Durbin1973}
Durbin J (1973).
newblock enquote{Weak convergence of the sample distribution function when
parameters are estimated.}
newblock emph{The Annals of Statistics}, textbf{1 (2)}, 279--290.
bibitem[{Durbin(1975)}]{Durbin1975}
Durbin J (1975).
newblock enquote{Kolmogorov-Smirnov Tests when Parameters are Estimated with
Applications to Tests of Exponentiality and Tests on Spacings.}
newblock emph{Biometrica}, textbf{62(1)}, 5--22.
bibitem[{Evans emph{et~al.}(1989)Evans, Johnson, and Green}]{Evans1989}
Evans JW, Johnson RA, Green DW (1989).
newblock emph{Two-and three-parameter weibull goodness-of-fit tests}, volume
newblock US Department of Agriculture, Forest Service, Forest Products
Laboratory.
bibitem[{Garfinkel and Mann(1968)}]{GarfinkelMann1968}
Garfinkel S, Mann W (1968).
newblock enquote{A method for obtaining large numbers of measured time
intervals in radioactive decay.}
newblock emph{International Journal of Applied Radiation and Isotopes,},
textbf{19(9)}, 707--709.
bibitem[{Geiger and Rutherford(1910)}]{GeigerRutherford1910}
Geiger H, Rutherford E (1910).
newblock enquote{The number of $alpha$ particles emitted by uranium and
thorium and by uranium minerals.}
newblock emph{Philolophical Magazine}, textbf{20(118)}, 691--698.
bibitem[{Kendall and Stuart(1979)}]{Kendall1979}
Kendall MG, Stuart A (1979).
newblock emph{The advanced theory of statistics - Inference and
relationship}, volume~2.
newblock 4th revised edition edition. C. Griffin, London.
newblock ISBN 978---0-8-52-64-2 -255- 2.
bibitem[{Kreer emph{et~al.}(2015)Kreer, Kizilersu, Thomas, and dos
Reis}]{KreerKizilersuThomasReis2015}
Kreer M, Kizilersu A, Thomas AW, dos Reis AE (2015).
newblock enquote{Goodness-of-fit tests and applications for left-truncated
Weibull distributions to non-life insurance.}
newblock emph{European Actuarial Journal}, textbf{5 (1)}.
newblock doi{10.1007/s13385-015-0105-8}.
bibitem[{Lehmann and Casella(1998)}]{LehmannCasella1998}
Lehmann E, Casella G (1998).
newblock emph{Theory of Point Estimation -}.
newblock 2nd ed. 1998. corr. 4th printing 2003 edition. Springer, Berlin,
Heidelberg.
newblock ISBN 978-0-387-98502-2.
bibitem[{Lilliefors(1969)}]{Lilliefors1969}
Lilliefors H (1969).
newblock enquote{On the Kolmogorov-Smirnov Test for the Exponential
Distribution with Mean Unknown.}
newblock emph{Journal of the American Statistical Association}, textbf{64
(325)}, 387--389.
bibitem[{Littell emph{et~al.}(1979)Littell, McClave, and Offen}]{Littell1979}
Littell RC, McClave JT, Offen WW (1979).
newblock enquote{Goodness-of-fit tests for the two parameter Weibull
distribution.}
newblock emph{Communications in Statististics - Simulation and Computation},
textbf{8 (3)}, 257--269.
bibitem[{Malevergne emph{et~al.}(2005)Malevergne, Pisarenko, and
Sornette}]{Malvergne2005}
Malevergne Y, Pisarenko V, Sornette D (2005).
newblock enquote{Empirical distributions of stock returns: between the
stretched exponential and the power law?}
newblock emph{Quantitative Finance}, textbf{5 (4)}, 379--401.
bibitem[{Massey(1951)}]{Massey1951}
Massey FJ (1951).
newblock enquote{The Kolmogorov-Smirnov Test for Goodness of Fit.}
newblock emph{Journal of the American Statistical Association},
textbf{46}(253), 68--78.
bibitem[{Miller(1956)}]{Miller1956}
Miller LH (1956).
newblock enquote{Table of Percentage Points of Kolmogorov Statistics.}
newblock emph{Journal of the American Statistical Association},
textbf{51}(273), pp. 111--121.
bibitem[{Parsons and Wirsching(1982)}]{ParsonsWirsching1982}
Parsons F, Wirsching PH (1982).
newblock enquote{A Kolmogorov-Smirnov goodness-of-fit test for the
two-parameter Weibull distribution when the parameters are estimated from the
data.}
newblock emph{Microelectronics Reliability}, textbf{22(2)}, 163--167.
bibitem[{Rinne(2009)}]{Rinne2009}
Rinne H (2009).
newblock emph{The Weibull Distribution: A Handbook}.
newblock 1st edition. CRC Press, Taylor & Francis Group, Boca Raton New York.
newblock ISBN 978-1-4200-8743-7.
bibitem[{Shorack and Wellner(2009)}]{Shorack2009}
Shorack GR, Wellner JA (2009).
newblock emph{Empirical processes with applications to statistics}.
newblock SIAM, Philadelphia.
newblock ISBN 978-0-898-71901-7.
bibitem[{Smirnov(1948)}]{Smirnov1948}
Smirnov N (1948).
newblock enquote{Table for Estimating the Goodness of Fit of Empirical
Distributions.}
newblock emph{The Annals of Mathematical Statistics}, textbf{19(2)},
--281.
bibitem[{Stephens(1977)}]{Stephens1977}
Stephens M (1977).
newblock enquote{Goodness of Fit for the Extreme Value Distribution.}
newblock emph{Biometrika}, textbf{64 (3)}, 583--588.
bibitem[{Thoman emph{et~al.}(1969)Thoman, Bain, and Antle}]{Thoman1969}
Thoman DR, Bain LJ, Antle CE (1969).
newblock enquote{Inferences on the parameters of the Weibull distribution.}
newblock emph{Technometrics}, textbf{11 (3)}, 445--460.
bibitem[{Weibull(1951)}]{Weibull1951}
Weibull W (1951).
newblock enquote{A statistical distribution function of wide applicability.}
newblock emph{J. Appl. Mech.-Trans. ASME}, textbf{73}, 293.
bibitem[{Wingo(1989)}]{Wingo1989}
Wingo DR (1989).
newblock enquote{The left-truncated Weibull distribution: theory and
computation.}
newblock emph{Statistical Papers}, textbf{30}, 39--48.
bibitem[{Woodruff emph{et~al.}(1983)Woodruff, Moore, Dunne, and
Cortes}]{Woodruff1983}
Woodruff BW, Moore AH, Dunne EJ, Cortes R (1983).
newblock enquote{A modified Kolmogorov-Smirnov test for Weibull distributions
with unknown location and scale parameters.}
newblock emph{IEEE Transactions on Reliability}, textbf{R-32 (2)}, 209--213.
Downloads
Additional Files
Published
How to Cite
Issue
Section
License
The Austrian Journal of Statistics publish open access articles under the terms of the Creative Commons Attribution (CC BY) License.
The Creative Commons Attribution License (CC-BY) allows users to copy, distribute and transmit an article, adapt the article and make commercial use of the article. The CC BY license permits commercial and non-commercial re-use of an open access article, as long as the author is properly attributed.
Copyright on any research article published by the Austrian Journal of Statistics is retained by the author(s). Authors grant the Austrian Journal of Statistics a license to publish the article and identify itself as the original publisher. Authors also grant any third party the right to use the article freely as long as its original authors, citation details and publisher are identified.
Manuscripts should be unpublished and not be under consideration for publication elsewhere. By submitting an article, the author(s) certify that the article is their original work, that they have the right to submit the article for publication, and that they can grant the above license.