Model selection in incomplete data
نویسنده
چکیده
Model selection in complete data is a common task for the applied researcher. However, in many scenarios data are incomplete which further complicates the task of model selection. In this talk, we will specify the problem of model selection in incomplete data and discuss several possible solutions using multiple imputation. First, we will define a new general measure for the correct model selection rates of common model selection criteria. Next, we will demonstrate the use of partial F-tests and define some new measures for model selection based on information criteria in multiply imputed data sets. This is a joint work with Ashok Chaurasia.
منابع مشابه
Selection of Variables that Influence Drug Injection in Prison: Comparison of Methods with Multiple Imputed Data Sets
Background: Prisoners, compared to the general population, are at greater risk of infection. Drug injection is the main route of HIV transmission, in particular in Iran. What would be of interest is to determine variables that govern drug injection among prisoners. However, one of the issues that challenge model building is incomplete national data sets. In this paper, we addressed the process ...
متن کاملFEM Updating for Offshore Jacket Structures Using Measured Incomplete Modal Data
Marine industry requires continued development of new technologies in order to produce oil. An essential requirement in design is to be able to compare experimental data from prototype structures with predicted information from a corresponding analytical finite element model. In this study, structural model updating may be defined as the fit of an existing analytical model in the light of measu...
متن کاملComplete / Incomplete Hierarchical Hub Center Single Assignment Network Problem
In this paper we present the problem of designing a three level hub center network. In our network, the top level consists of a complete network where a direct link is between all central hubs. The second and third levels consist of star networks that connect the hubs to central hubs and the demand nodes to hubs and thus to central hubs, respectively. We model this problem in an incomplete net...
متن کاملStochastic complexity and model selection from incomplete data
The principle of minimum description length (MDL) provides an approach for selecting the model class with the smallest stochastic complexity of the data among a set of model classes. However, when only incomplete data are available the stochastic complexity for the complete data cannot be numerically computed. In this paper, this problem is solved by introducing a notion of expected stochastic ...
متن کاملFormal and Informal Model Selection with Incomplete Data
Model selection and assessment with incomplete data pose challenges in addition to the ones encountered with complete data. There are two main reasons for this. First, many models describe characteristics of the complete data, in spite of the fact that only an incomplete subset is observed. Direct comparison between model and data is then less than straightforward. Second, many commonly used mo...
متن کامل