Alonso Revenga, Juana MaríaMartín Apaolaza, NirianPardo Llorente, LeandroBrugnano, LuigiEfendiev, YalchinKeller, AndréKwok-Po, MichaelRomani, LuciaTank. Fatih2024-05-292024-05-292020-08-15Alonso-Revenga, Martín y Pardo (2020) «New statistics to test log-linear modeling hypothesis with no distributional specifications and clusters with homogeneous correlation», Journal of Computational and Applied Mathematics, 374.0377-042710.1016/J.CAM.2020.112757https://hdl.handle.net/20.500.14352/104517Traditionally, the Dirichlet-multinomial distribution has been recognized as a key model for contingency tables generated by cluster sampling schemes. There are, however, other possible distributions appropriate for these contingency tables. This paper introduces new statistics capable of testing log-linear modeling hypotheses with distributional unspecification, when the individuals of the clusters are possibly homogeneously correlated. An estimator for the intracluster correlation coefficient, valid for different cluster sizes, plays a crucial role in the construction of the goodness-of-fit test-statistics.engAttribution-NonCommercial-NoDerivatives 4.0 Internationaljournal article1879-1778https://doi.org/10.1016/j.cam.2020.112757https://www.sciencedirect.com/science/article/pii/S0377042720300480?via%3Dihubrestricted access519.235004.415.26Clustered multinomial dataConsistent intracluster correlationEstimatorLog-linear modelOverdispersionQuasi minimum divergence estimatorProbabilidades (Estadística)1208 Probabilidad