Ignorability for categorical data. (English) Zbl 1078.62002
Summary: We study the problem of ignorability in likelihood-based inference from incomplete categorical data. Two versions of the coarsened at random assumption (car) are distinguished, their compatibility with the parameter distinctness assumption is investigated and several conditions for ignorability that do not require an extra parameter distinctness assumption are established.
It is shown that car assumptions have quite different implications depending on whether the underlying complete-data model is saturated or parametric. In the latter case, car assumptions can become inconsistent with observed data.
It is shown that car assumptions have quite different implications depending on whether the underlying complete-data model is saturated or parametric. In the latter case, car assumptions can become inconsistent with observed data.
MSC:
| 62A01 | Foundations and philosophical topics in statistics |
| 62N01 | Censored data models |
| 62H17 | Contingency tables |
Keywords:
coarse data; contingency tables; maximum likelihood inference; missing at random; missing valuesReferences:
| [1] | Cator, E. (2004). On the testability of the CAR assumption. Ann. Statist. 32 1957–1980. · Zbl 1056.62003 · doi:10.1214/009053604000000418 |
| [2] | Cowell, R. G., Dawid, A. P., Lauritzen, S. L. and Spiegelhalter, D. J. (1999). Probabilistic Networks and Expert Systems . Springer, New York. · Zbl 0937.68121 · doi:10.1007/b97670 |
| [3] | Darroch, J., Lauritzen, S. L. and Speed, T. (1980). Markov fields and log-linear interaction models for contingency tables. Ann. Statist. 8 522–539. · Zbl 0444.62064 · doi:10.1214/aos/1176345006 |
| [4] | Dawid, A. P. and Dickey, J. M. (1977). Likelihood and Bayesian inference from selectively reported data. J. Amer. Statist. Assoc. 72 845–850. · Zbl 0372.62025 · doi:10.2307/2286472 |
| [5] | Dempster, A. P. (1967). Upper and lower probabilities induced by a multivalued mapping. Ann. Math. Statist. 38 325–339. · Zbl 0168.17501 · doi:10.1214/aoms/1177698950 |
| [6] | Dempster, A. P., Laird, N. M. and Rubin, D. B. (1977). Maximum likelihood from incomplete data via the EM algorithm (with discussion). J. Roy. Statist. Soc. Ser. B 39 1–38. · Zbl 0364.62022 |
| [7] | Gill, R. D., van der Laan, M. J. and Robins, J. M. (1997). Coarsening at random: Characterizations, conjectures, counter-examples. Proc. First Seattle Symposium in Bio- statistics : Survival Analysis . Lecture Notes in Statist. 123 255–294. Springer, New York. · Zbl 0918.62003 |
| [8] | Grünwald, P. and Halpern, J. (2003). Updating probabilities. J. Artificial Intelligence Res. 19 243–278. · Zbl 1076.68579 |
| [9] | Heitjan, D. F. (1994). Ignorability in general incomplete-data models. Biometrika 81 701–708. · Zbl 0810.62008 · doi:10.1093/biomet/81.4.701 |
| [10] | Heitjan, D. F. (1997). Ignorability, sufficiency and ancillarity. J. Roy. Statist. Soc. Ser. B 59 375–381. · Zbl 0886.62006 · doi:10.1111/1467-9868.00073 |
| [11] | Heitjan, D. F. and Rubin, D. B. (1991). Ignorability and coarse data. Ann. Statist. 19 2244–2253. JSTOR: · Zbl 0745.62004 · doi:10.1214/aos/1176348396 |
| [12] | Jordan, M. I., ed. (1999). Learning in Graphical Models . MIT Press, Cambridge, MA. · Zbl 0889.00024 |
| [13] | Lauritzen, S. L. and Spiegelhalter, D. J. (1988). Local computations with probabilities on graphical structures and their application to expert systems (with discussion). J. Roy. Statist. Soc. Ser. B 50 157–224. · Zbl 0684.68106 |
| [14] | McLachlan, G. and Krishnan, T. (1997). The EM Algorithm and Extensions. Wiley, New York. · Zbl 0882.62012 |
| [15] | Rubin, D. (1976). Inference and missing data (with discussion). Biometrika 63 581–592. · Zbl 0344.62034 · doi:10.1093/biomet/63.3.581 |
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.
