×

A mixture model for occupational exposure mean testing with a limit of detection. (English) Zbl 1209.62234

Summary: Information from detectable exposure measurements randomly sampled from a left-truncated log-normal distribution may be used to evaluate the distribution of nondetectable values that fall below an analytic limit of detection. If the proportion of nondetects is larger than expected under log normality, alternative models to account for these unobserved data should be considered. We discuss one such model that incorporates a mixture of true zero exposures and a log-normal distribution with possible left censoring, previously considered in a different context by L. H. Moulton and N. A. Halsey [Biometrics 51, No. 4, 1570–1578 (1995; Zbl 0875.62502)]. A particular relationship is demonstrated between maximum likelihood parameter estimates based on this mixture model and those assuming either left-truncated or left-censored data. These results emphasize the need for caution when choosing a model to fit data involving nondetectable values. A one-sided likelihood ratio test for comparing mean exposure under the mixture model to an occupational exposure limit is then developed and evaluated via simulations. An example demonstrates the potential impact of specifying an incorrect model for the nondetectable values.

MSC:

62N03 Testing in survival analysis and censored data
62N01 Censored data models
62F10 Point estimation
62F03 Parametric hypothesis testing
65C60 Computational problems in statistics (MSC2010)

Citations:

Zbl 0875.62502

Software:

SAS/IML
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Aitchison, On the distribution of a positive random variable having a discrete probability mass at the origin, Journal of the American Statistical Association 50 pp 901– (1955) · Zbl 0065.12604 · doi:10.2307/2281175
[2] Bar-Shalom, Handbook of Statistical Tests for Evaluating Employee Exposure to Air Contaminants pp 75– (1975)
[3] Boag, Maximum likelihood estimation of the proportion of patients cured by cancer therapy, Journal of the Royal Statistical Society, Series B 11 pp 15– (1949) · Zbl 0034.08001
[4] Brown, The small sample behavior of some statistics with heterogeneous variances, Technometrics 30 pp 129– (1974) · Zbl 0275.62047 · doi:10.2307/1267501
[5] Cohen, Simplified estimators for the normal distribution when samples are singly censored or truncated, Technometrics 1 pp 217– (1959) · Zbl 0201.52003 · doi:10.2307/1266442
[6] Dempster, Maximum likelihood from incomplete data via the EM algorithm, Journal of the Royal Statistical Society, Series B 39 pp 1– (1977) · Zbl 0364.62022
[7] El-Shaarawi, Inferences about the mean from censored water quality data, Water Resources Research 25 pp 685– (1989) · doi:10.1029/WR025i004p00685
[8] Farewell, The use of mixture models for the analysis of survival data with long-term survivors, Biometrics 38 pp 1041– (1982) · doi:10.2307/2529885
[9] Fears, A reminder of the fallibility of the Wald statistic, The American Statistician 50 pp 226– (1996) · Zbl 04535665 · doi:10.2307/2684659
[10] Ghitany, Estimating the proportion of immunes in censored samples: A simulation study, Statistics in Medicine 14 pp 39– (1995) · doi:10.1002/sim.4780140106
[11] Gilbert, Statistical Methods for Environmental Pollution Monitoring (1987)
[12] Gill, Procedures for optimization problems with a mixture of bounds and general linear constraints, Mathematical Software 10 pp 282– (1984) · Zbl 0562.90079 · doi:10.1145/1271.1276
[13] Gilliom, Estimation of distributional parameters for censored trace level water quality data. 1. Estimation techniques, Water Resources Research 22 pp 135– (1986) · doi:10.1029/WR022i002p00135
[14] Goldman, Survivorship analysis when cure is a possibility: A Monte Carlo study, Statistics in Medicine 3 pp 153– (1984) · doi:10.1002/sim.4780030208
[15] Hewett, Mean testing. II. Comparison of several alternative procedures, Applied Occupational and Environmental Hygiene 12 pp 347– (1997) · doi:10.1080/1047322X.1997.10389517
[16] Hewett, Simple procedures for calculating confidence intervals around sample mean and exceedance fractions derived from lognormally distributed data, Applied Occupational and Environmental Hygiene 12 pp 132– (1997) · doi:10.1080/1047322X.1997.10389473
[17] Hinton, {\(\times\)} log-normal statistical methodology performance, Environmental Science Technology 27 pp 2247– (1993) · doi:10.1021/es00047a036
[18] Kromhout, A comprehensive evaluation of within- and between-worker components of occupational exposure to chemical agents, Annals of Occupational Hygiene 37 pp 253– (1993) · doi:10.1093/annhyg/37.3.253
[19] Land, Confidence intervals for linear functions of the normal mean and variance, Annals of Mathematical Statistics 42 pp 1187– (1971) · Zbl 0223.62046 · doi:10.1214/aoms/1177693235
[20] Land, Selected Tables in Mathematical Statistics Volume III pp 385– (1975)
[21] Lange, Numerical Analysis For Statisticians (1999) · Zbl 0920.62001
[22] Lyles, On strategies for comparing occupational exposure data to limits, American Industrial Hygiene Association Journal 57 pp 6– (1996) · doi:10.1080/15428119691015430
[23] Maller, Estimating the proportion of immunes in a censored sample, Biometrika 79 pp 731– (1992) · Zbl 0764.62094 · doi:10.1093/biomet/79.4.731
[24] Maller, Testing for the presence of immune or cured individuals in censored survival data, Biometrics 51 pp 1197– (1995) · Zbl 0875.62516 · doi:10.2307/2533253
[25] Meeker, Teaching about approximate confidence regions based on maximum likelihood estimation, The American Statistician 49 pp 48– (1995) · doi:10.2307/2684811
[26] Moulton, A mixture model with detection limits for regression analysis of antibody response to vaccine, Biometrics 51 pp 1570– (1995) · Zbl 0875.62502 · doi:10.2307/2533289
[27] Oldham, The nature of the variability of dust concentrations at the coal face, British Journal of Industrial Medicine 10 pp 227– (1953)
[28] Owen, Estimation of the mean for lognormal data containing zeroes and left-censored values, with applications to the measurement of worker exposure to air contaminants, Biometrics 36 pp 707– (1980) · doi:10.2307/2556125
[29] Paarsch, A Monte Carlo comparison of estimators for censored regression models, Journal of Econometrics 24 pp 197– (1984) · Zbl 0527.62100 · doi:10.1016/0304-4076(84)90079-4
[30] Powell, Algorithms for nonlinear constraints that use Lagrangian functions, Mathematical Programming 14 pp 224– (1978) · Zbl 0383.90092 · doi:10.1007/BF01588967
[31] Rappaport, Assessment of long-term exposures to toxic substances in air, Annals of Occupational Hygiene 35 pp 61– (1991) · doi:10.1093/annhyg/35.1.61
[32] SAS Institute, SAS/IML Software: Changes and Enhancements through Release 6.11 (1995)
[33] Self, Asymptotic properties of maximum likelihood estimators and likelihood ratio tests under nonstandard conditions, Journal of the American Statistical Association 82 pp 605– (1987) · Zbl 0639.62020 · doi:10.2307/2289471
[34] Silvapulle, On tests of one-sided hypotheses in some generalized linear models, Biometrics 50 pp 853– (1994) · Zbl 0821.62041 · doi:10.2307/2532799
[35] Spear, Benzene exposure in the petroleum refining industry, Applied Industrial Hygiene 2 pp 155– (1987) · doi:10.1080/08828032.1987.10390543
[36] Welch, On the comparison of several mean values, Biometrika 38 pp 330– (1951) · Zbl 0043.14101 · doi:10.1093/biomet/38.3-4.330
[37] Zhou, Comparison of several independent population means when their samples contain log-normal and possibly zero observations, Biometrics 55 pp 645– (1999) · Zbl 1059.62518 · doi:10.1111/j.0006-341X.1999.00645.x
[38] Zhou, Confidence intervals for the mean of diagnostic test charge data containing zeros, Biometrics 56 pp 1118– (2000) · Zbl 1060.62686 · doi:10.1111/j.0006-341X.2000.01118.x
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.