Adjustments for reporting delays and the prediction of occurred but not reported events

We consider delays that occur in the reporting of events such as cases of a reportable disease or insurance claims. Estimation of the number of events that have occurred but not yet been reported (OBNR events) is then important. Current methods of doing this do not allow random temporal fluctuations in reporting delays, and consequently, confidence or prediction limits on OBNR events tend to be too narrow. We develop an approach that uses recent reporting data and incorporates random effects, thus leading to more reasonable and robust predictions Nous considerons les delais qui ont lieu dans le rapport d'evenements tels qu'une maladie ou une demande d'indemnite. L'estimation du nombre d'evenements qui ont eu lieu mais n'ont pas encore ete rapportes (evenements OBNR) est alors importante. Les methodes actuelles ne permettent pas de fluctuations temporelles aleatoires dans les delais de rapport et par consequent, les limites de confiance ou de prevision des evenements OBNR ont tendance a etre trop etroites. Nous developpons une approche utilisant les donnees de rapport recent et incorporant des effets aleatoires. Ainsi, les previsions sont plus vraisemblables et robustes.

[1]  Morgan Wm,et al.  Acquired immunodeficiency syndrome: current and future trends. , 1986, Public health reports.

[2]  R Brookmeyer,et al.  Statistical methods for short-term projections of AIDS incidence. , 1989, Statistics in medicine.

[3]  Jeffrey E. Harris,et al.  Reporting Delays and the Incidence of AIDS , 1990 .

[4]  P. Holland,et al.  Discrete Multivariate Analysis. , 1976 .

[5]  V. De Gruttola,et al.  Nonparametric analysis of truncated survival data, with application to AIDS , 1988 .

[6]  H. White Maximum Likelihood Estimation of Misspecified Models , 1982 .

[7]  N. Inagaki Asymptotic relations between the likelihood estimating function and the maximum likelihood estimator , 1973 .

[8]  D. Murdoch,et al.  The probability of degenerate maximum-likelihood estimates with randomly truncated survival data , 1993 .

[9]  R Brookmeyer,et al.  The analysis of delays in disease reporting: methods and results for the acquired immunodeficiency syndrome. , 1990, American journal of epidemiology.

[10]  N. L. Johnson,et al.  Distributions in Statistics: Discrete Distributions. , 1970 .

[11]  J. Kalbfleisch,et al.  The Statistical Analysis of Failure Time Data , 1980 .

[12]  Kenneth S. Kaminsky,et al.  Prediction of IBNR claim counts by modelling the distribution of report lags , 1987 .

[13]  Greg Taylor Claims reserving in non-life insurance , 1985 .

[14]  N. Jewell,et al.  Uncertainty About the Incubation Period of AIDS and Its Impact on Backcalculation , 1992 .

[15]  Owen Devine,et al.  Predicting AIDS incidence by extrapolating from recent trends , 1989 .

[16]  A Credibility Model with Random Fluctuations in Delay Probabilities for the Prediction of IBNR Claims ( , 1988 .

[17]  J B Carlin,et al.  A method of non-parametric back-projection and its application to AIDS data. , 1991, Statistics in medicine.

[18]  Jerald F. Lawless,et al.  Inference Based on Retrospective Ascertainment: An Analysis of the Data on Transfusion-Related AIDS , 1989 .

[19]  John D. Kalbfleisch,et al.  Methods for the analysis and predic tion of warranty claims , 1991 .