Incomplete Contingency Research Articles

On log-linear modeling for an incomplete two-way contingency table with one variable subject to nonresponse

In this paper we address two issues in the use of the nonresponse log-linear models for the analysis of an incomplete two-way contingency table with one variable subject to nonresponse, the occurrence of nonresponse boundary solutions and the assessment of the missing data mechanism. To this end, we employ a set of response odds from the fully classified counts and nonresponse odds from partially classified counts. We first investigate the role of the set of odds in identifying the occurrence of boundary solutions and assessing the nonresponse log-linear models suitable for the data, and then propose a data analytic guideline for the analysis of an incomplete two-way contingency table. We also examine the theoretical properties of the set of odds when nonresponse boundary solutions occur.

Large tests of independence in incomplete two-way contingency tables using fractional imputation

Imputation procedures fill-in missing values, thereby enabling complete data analyses. Fully efficient fractional imputation (FEFI) and multiple imputation (MI) create multiple versions of the missing observations, thereby reflecting uncertainty about their true values. Methods have been described for hypothesis testing with multiple imputation. Fractional imputation assigns weights to the observed data to compensate for missing values. The focus of this article is the development of tests of independence using FEFI for partially classified two-way contingency tables. Wald and deviance tests of independence under FEFI are proposed. Simulations are used to compare type I error rates and Power. The partially observed marginal information is useful for estimating the joint distribution of cell probabilities, but it is not useful for testing association. FEFI compares favorably to other methods in simulations.

Fractional imputation for incomplete two-way contingency tables

Imputation procedures such as fully efficient fractional imputation (FEFI) or multiple imputation (MI) create multiple versions of the missing observations, thereby reflecting uncertainty about their true values. Multiple imputation generates a finite set of imputations through a posterior predictive distribution. Fractional imputation assigns weights to the observed data. The focus of this article is the development of FEFI for partially classified two-way contingency tables. Point estimators and variances of FEFI estimators of population proportions are derived. Simulation results, when data are missing completely at random or missing at random, show that FEFI is comparable in performance to maximum likelihood estimation and multiple imputation and superior to simple stochastic imputation and complete case anlaysis. Methods are illustrated with four data sets.

Bayesian analysis for incomplete multi-way contingency tables with nonignorable nonresponse

We propose Bayesian methods with five types of priors to estimate cell probabilities in an incomplete multi-way contingency table under nonignorable nonresponse. In this situation, the maximum likelihood (ML) estimates often fall in the boundary solution, causing the ML estimates to become unstable. To deal with such a multi-way table, we present an EM algorithm which generalizes the previous algorithm used for incomplete one-way tables. Three of the five types of priors were previously introduced while the other two are newly proposed to reflect different response patterns between respondents and nonrespondents. Data analysis and simulation studies show that Bayesian estimates based on the old three priors can be worse than the ML regardless of occurrence of boundary solution, contrary to previous studies. The Bayesian estimates from the two new priors are most preferable when a boundary solution occurs. We provide an illustrating example using data for a study of the relationship between a mother's smoking and her newborn's weight.

A Multilevel Model for Continuous Time Population Estimation

Statistical methods have been developed and applied to estimating populations that are difficult or too costly to enumerate. Known as multilist methods in epidemiological settings, individuals are matched across lists and estimation of population size proceeds by modeling counts in incomplete multidimensional contingency tables (based on patterns of presence/absence on lists). As multilist methods typically assume that lists are compiled instantaneously, there are few options available for estimating the unknown size of a closed population based on continuously (longitudinally) compiled lists. However, in epidemiological settings, continuous time lists are a routine byproduct of administrative functions. Existing methods are based on time-to-event analyses with a second step of estimating population size. We propose an alternative approach to address the twofold epidemiological problem of estimating population size and of identifying patient factors related to duration (in days) between visits to a health care facility. A Bayesian framework is proposed to model interval lengths because, for many patients, the data are sparse; many patients were observed only once or twice. The proposed method is applied to the motivating data to illustrate the methods' applicability. Then, a small simulation study explores the performance of the estimator under a variety of conditions. Finally, a small discussion section suggests opportunities for continued methodological development for continuous time population estimation.

Markov chain Monte Carlo exact tests for incomplete two-way contingency tables

We consider testing the quasi-independence hypothesis for two-way contingency tables which contain some structural zero cells. For sparse contingency tables where the large sample approximation is not adequate, the Markov chain Monte Carlo exact tests are powerful tools. To construct a connected chain over the two-way contingency tables with fixed sufficient statistics and an arbitrary configuration of structural zero cells, an algebraic algorithm proposed by Diaconis and Sturmfels [Diaconis, P. and Sturmfels, B. (1998). The Annals of statistics, 26, pp. 363–397.] can be used. However, their algorithm does not seem to be a satisfactory answer, because the Markov basis produced by the algorithm often contains many redundant elements and is hard to interpret. We derive an explicit characterization of a minimal Markov basis, prove its uniqueness, and present an algorithm for obtaining the unique minimal basis. A computational example and the discussion on further basis reduction for the case of positive sufficient statistics are also given.

Minimum chi-square estimator and test for incomplete multidimensional contingency tables

A modified transformed chi-square statistic is defined for testing hypotheses of quasi-independence in the incomplete multi-dimensional contingency table and a simple method for determining degrees of freedom is given. A modified transformed chi-squareestimator of the expected cell frequencies is given in closed form for a general class of exact linear constraints. The co-variance matrix of estimated cell frequencies is derived under the assumption of a conditional Poisson distribution.

The Analysis of Contingency Tables with Incompletely Classified Data

In many practical situations, investigators are forced to study the structure underlying the crossclassification of several categorical variables via tables of observed counts in which the observations corresponding to certain sets of cells are indistinguishable. Methods are presented for the analysis of such contingency tables with incompletely cross-classified data via loglinear models. The method of maximum likelihood is used to estimate the expected cell counts which are then used to test the goodness-of-fit of the model. Extensions to incomplete (or truncated) contingency tables are indicated and several examples are given.

Childhood Mortality, Family Size and Birth Order in Pre-Industrial Europe

Based on parish registers, demographic histories of Crulai (France), Tourouvre-au-Perche (France), and Geneva (Swizertland) established the childhood mortality experienced by complete sibships during periods of at least half a century before the Fench revolution. These observations may be presented as frequenceis in incomplete five-dimensional contingency tables. The five dimensions are: survival (living or dead), completed sibship size, birth order, type of family (according to completeness of information about family), and epoch (period in which the family lived). This paper reanalyzes these published data, using hierarchical log-linear models to discern which interations among the five variables can justifiably be inferred from the data. The neonatal and infant mortality rates of firstborn are probably higher than those of later sibs (in Crulai and Tourouvre). But mortality by age 20 (in Geneva) is associated strongly with the epoch, type of family, and family size, and not significantly with birth order. The increase in mortality with completed family size is insufficient to select, in an evolutionary sense, for limited family size.

Correction: The Analysis of Incomplete Multi-Way Contingency Tables

Correction: The Analysis of Incomplete Multi-Way Contingency Tables

Testing the equality of risk difference among multiple incomplete two-way contingency tables

Testing the equality of risk difference among multiple incomplete two-way contingency tables