×

Estimation of upper quantiles under model and parameter uncertainty. (English) Zbl 1101.62318

Summary: In this paper we assess accuracy of some commonly used estimators of upper quantiles of a right skewed distribution under both parameter and model uncertainty. In particular, for each of log-normal, log-logistic, and log-double exponential distributions, we study the bias and mean squared error of the maximum likelihood estimator (MLE) of the upper quantiles under both the correct and incorrect model specifications. We also consider two data dependent or adaptive estimators. The first (tail-exponential) is based on fitting an exponential distribution to the highest 10–20 percent of the data. The second selects the best fitting likelihood-based model and uses the MLE obtained from that model. The simulation results provide some practical guidance concerning the estimation of the upper quantiles when one is uncertain about the underlying model. We found that the consequences of assuming log-normality when the true distribution is log-logistic or log-double exponential are not severe in moderate sample sizes. For extreme quantiles, no estimator was reliable in small samples. For large sample sizes the selection estimator performs fairly well. For small sample sizes the tail-exponential method is a good alternative. Presenting it and the MLE for the log-normal enables one to assess the potential effects of model uncertainty.

MSC:

62F10 Point estimation
65C05 Monte Carlo methods

Software:

bootstrap; EnvStats
Full Text: DOI

References:

[1] Aitchison, J.; Brown, J. A. C.: The lognormal distribution. (1973) · Zbl 0081.14303
[2] Aitchison, A. C.: Regression diagnostics, transformations and constructed variables. J. roy. Statist. soc. Ser. B 1, 1-36 (1982) · Zbl 0508.62058
[3] Andrews, D. F.: Robust estimation of location. (1972) · Zbl 0254.62001
[4] Bennett, S.: Log-logistic regression models for survival data. Appl. statist. 32, No. 2, 165-171 (1983)
[5] Breiman, L., Gins, J., Stone, C., 1979. New Methods for Estimating Tail Probabilities and Extreme Value Distributions. Technology Service Corp., Santa Monica, CA, TSC-PD-A2261.
[6] Burmaster, D. E.: Lognormal distribution for skin area as a function of body weight. Risk anal. 18, No. 1, 27-32 (1998)
[7] Burmaster, D. E.; Crouch, A. C.: Lognormal distributions for body weight as a function of age for males and females in the united states, 1976–1980. Risk anal. 17, No. 4, 499-505 (1997)
[8] D’agostino, R. B.; Stephens, M. A.: Goodness-of-fit techniques. (1986) · Zbl 0597.62030
[9] Draper, D.: Assessment and propagation of model uncertainty. J. R. Statist. soc. Ser. B 57, No. 1, 45-97 (1995) · Zbl 0812.62001
[10] Dubey, S. D.: Transformation for estimation of parameters. J. ind. Statist. assoc. 4, 109-124 (1966)
[11] Dumonceaux, R.; Antle, C. E.: Likelihood ratio test for discrimination between two models with unknown location and scale parameters. Technometrics 15, No. 1, 19-27 (1973) · Zbl 0264.62008
[12] Dyer, A. R.: Discrimination procedures for separate families of hypotheses. J. amer. Statist. assoc. 68, 970-974 (1973) · Zbl 0271.62073
[13] Efron, B.; Tibshirani, R. J.: An introduction to bootstrap. (1993) · Zbl 0835.62038
[14] Frey, H. C.; Burmaster, D. E.: Methods for characterizing variability and uncertainty: comparison of bootstrap and likelihood-based approaches. Risk anal. 19, No. 1, 109-130 (1999)
[15] Firth, D.: Multiplicative errors: log-normal or gamma?. J. R. Statist. soc. B 50, No. 2, 266-268 (1988)
[16] Gastwirth, J. L.: On robust procedures. J. amer. Statist. assoc. 61, No. 316, 929-948 (1966) · Zbl 0144.19004
[17] Haas, C. N.: Importance of distributional form in characterizing inputs to Monte Carlo risk assessment. Risk anal. 17, No. 1, 107-113 (1997)
[18] Hoaglin, D.C., Mosteller, F., Tukey, J.W. (Eds.), 1983. Understanding Robust and Exploratory Data Analysis. Wiley, New York. · Zbl 0599.62007
[19] Johnson, N. L.: Systems of frequency curves generated by methods of translation. Biometrika 36, 149-176 (1949) · Zbl 0033.07204
[20] Johnson, N. L.; Kotz, S.; Balakrishnan, N.: Continuous univariate distributions, volumes I and II. (1995) · Zbl 0821.62001
[21] Kappenman, R. F.: On a method for selecting a distributional model. Commun. statist. Theory methods 11, No. 6, 663-672 (1982)
[22] Kapteyn, J. C.: Skew frequency curves in biology and statistics. Astronomical laboratory. (1903) · JFM 34.0268.03
[23] Lehman, E. L.: Theory of point estimation. (1983)
[24] Mage, D. T.: A review of applications of probability models for describing aerometric data. Environmetrics 81: selected papers. (1981)
[25] Mccullagh, P.; Nelder, J. A.: Generalized linear models. (1989) · Zbl 0744.62098
[26] Millard, S. P.: Environmental stats for S-plus. (1998)
[27] Mudholkar, G. S.; George, E. O.: A remark on the shape of the logistic distribution. Biometrika 65, No. 3, 667-669 (1978) · Zbl 0389.62009
[28] Murray, D. M.; Burmaster, D. E.: Estimated distributions for average daily consumption of total and self-caught fish for adult michigan angler households. Risk anal. 14, No. 4, 513-519 (1994)
[29] Ott, W. R.: Environmental statistics and data analysis. (1995) · Zbl 0859.62106
[30] Quesenberry, C. P.; Kent, J.: Selecting among probability distributions used in reliability. Technometrics 24, No. 1, 59-66 (1982)
[31] Rosebury, A. M.; Burmaster, D. E.: Log-normal distributions for water intake for children and adults. Risk anal. 12, No. 1, 99-104 (1992)
[32] Rustagi, J. S.: Stochastic behavior of trace substances. Arch. environ. Health 8, 84-87 (1964)
[33] Séménou, M.: Quantile estimation under possibly misspecified generalized linear model. Statist. probab. Lett. 27, 357-365 (1996) · Zbl 0847.62059
[34] Tadikamalla, P. R.; Johnson, N. L.: Systems of frequency curves generated by transformations of logistic variables. Biometrika 69, No. 2, 461-465 (1982)
[35] Uppuluri, V.R.R., 1980. Some properties of log-laplace distribution. Manuscript No. 103, International Summer School on Statistical Distributions in Scientific Work Trieste, Italy, July-August. · Zbl 0476.62015
[36] Wiens, B. L.: When log-normal and gamma models give different results: a case study. Amer. statist. 53, No. 2, 89-93 (1999)
[37] USEPA, 1985. Technical Support Documents for Water Quality-Based Toxics Control. Washington DC.
[38] USEPA, 1992. Statistical analysis of ground-water monitoring data at RCRA facilities. Office of Solid Waste, Washington, DC, EPA/530-R-93-003.
[39] USEPA, 2001. Web site of the Office of Water at the United States Environmental Protection Agency: (www.epa.gov/safewater).
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.