Intrinsic Tests of Independence in Contingency Tables
نویسندگان
چکیده
For Bayesian testing of independence in contingency tables, the priors play an important role regardless of the sample size. The simplest default prior, the uniform prior on the simplex, may sometimes not give sensible answers. However, we show that intrinsic priors derived from the uniform greatly improve the performance of Bayesian tests. The intrinsic priors give more mass to alternative parameter values that are close to the null hypothesis, and less prior probability to extreme models, while maintaining consistency of the test. We also discuss conditioning issues and sampling schemes, and argue that conditioning should be on either one margin or the table total, but not on both margins. Examples using real and simulated data are given.
منابع مشابه
Assessing Robustness of Intrinsic Tests of Independence in Two-Way Contingency Tables
A condition needed for testing nested hypotheses from a Bayesian viewpoint is that the prior for the alternative model concentrates mass around the smaller, or null, model. For testing independence in contingency tables, the intrinsic priors satisfy this requirement. Further, the degree of concentration of the priors is controlled by a discrete parameter m, the training sample size, which plays...
متن کاملPlain Answers to Several Questions about Association/Independence Structure in Complete/Incomplete Contingency Tables
In this paper, we develop some results based on Relational model (Klimova, et al. 2012) which permits a decomposition of logarithm of expected cell frequencies under a log-linear type model. These results imply plain answers to several questions in the context of analyzing of contingency tables. Moreover, determination of design matrix and hypothesis-induced matrix of the model will be discusse...
متن کاملSpeeding up the execution of a large number of statistical tests of independence
A massive amount of conditional independence tests on data must be performed in the problem of learning the structure of probabilistic graphical models when using the independence-based approach. An intermediate step in the computation of independence tests is the construction of contingency tables from the data. In this work we present an intelligent cache of contingency tables that allows the...
متن کاملMultiple Testing in Large-Scale Contingency Tables: Inferring Pair-Wise Amino Acid Patterns in β-Sheets
One of the most common test procedures using two-way contingency tables is a test of independence between two categorizations. Current significant tests such as χ2 tests or likelihood ratio tests provide overall independency but bring limited information about the nature of the association in the contingency tables. This study examines the feasibility of using multiple testing procedures for an...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2006