Browsing by Subject "BIC"
Now showing 1 - 5 of 5
Results Per Page
Sort Options
Item Assigning g in Zellner's g prior for Bayesian variable selection(2015-05) Wang, Mengjie; Walker, Stephen G., 1945-; Lin, LizhenThere are numerous frequentist statistics variable selection methods such as Stepwise regression, AIC and BIC etc. In particular, the latter two criteria include a penalty term which discourages overfitting. In terms of the framework of Bayesian variable selection, a popular approach is using Bayes Factor (Kass & Raftery 1995), which also has a natural built-in penalty term (Berger & Pericchi 2001). Zellner's g prior (Zellner 1986) is a common prior for coefficients in the linear regression model due to its computational speed of analytic solutions for posterior. However, the choice of g is a problem which has attracted a lot of attention. (Zellner 1986) pointed out that if g is unknown, a prior can be introduced and g can be integrated out. One of the prior choices is Hyper-g Priors proposed by (Liang et al. 2008). Instead of proposing a prior for g, we will assign a fixed value for g based on controlling the Type I error for the test based on the Bayes factor. Since we are using Bayes factor to do model selection, the test statistic is Bayes factor. Every test comes with a Type I error, so it is reasonable to restrict this error under a critical value, which we will take as benchmark values, such as 0.1 or 0.05. This approach will automatically involve setting a value of g. Based on this idea, a fixed g can be selected, hence avoiding the need to find a prior for g.Item Bayesian model selection using exact and approximated posterior probabilities with applications to Star Data(Texas A&M University, 2004-11-15) Pokta, SurianiThis research consists of two parts. The first part examines the posterior probability integrals for a family of linear models which arises from the work of Hart, Koen and Lombard (2003). Applying Laplace's method to these integrals is not entirely straightforward. One of the requirements is to analyze the asymptotic behavior of the information matrices as the sample size tends to infinity. This requires a number of analytic tricks, including viewing our covariance matrices as tending to differential operators. The use of differential operators and their Green's functions can provide a convenient and systematic method to asymptotically invert the covariance matrices. Once we have found the asymptotic behavior of the information matrices, we will see that in most cases BIC provides a reasonable approximation to the log of the posterior probability and Laplace's method gives more terms in the expansion and hence provides a slightly better approximation. In other cases, a number of pathologies will arise. We will see that in one case, BIC does not provide an asymptotically consistent estimate of the posterior probability; however, the more general Laplace's method will provide such an estimate. In another case, we will see that a naive application of Laplace's method will give a misleading answer and Laplace's method must be adapted to give the correct answer. The second part uses numerical methods to compute the "exact" posterior probabilities and compare them to the approximations arising from BIC and Laplace's method.Item Factor Analysis for Skewed Data and Skew-Normal Maximum Likelihood Factor Analysis(2013-04-04) Gaucher, Beverly JaneThis research explores factor analysis applied to data from skewed distributions for the general skew model, the selection-elliptical model, the selection-normal model, the skew-elliptical model and the skew-normal model for finite sample sizes. In terms of asymptotics, or large sample sizes, quasi-maximum likelihood methods are broached numerically. The skewed models are formed using selection distribution theory, which is based on Rao?s weighted distribution theory. The models assume the observed variable of the factor model is from a skewed distribution by defining the distribution of the unobserved common factors skewed and the unobserved unique factors symmetric. Numerical examples are provided using maximum likelihood selection skew-normal factor analysis. The numerical examples, such as maximum likelihood parameter estimation with the resolution of the ?sign switching? problem and model fitting using likelihood methods, illustrate that the selection skew-normal factor analysis model better fits skew-normal data than does the normal factor analysis model.Item Testing Lack-of-Fit of Generalized Linear Models via Laplace Approximation(2012-07-16) Glab, Daniel LaurenceIn this study we develop a new method for testing the null hypothesis that the predictor function in a canonical link regression model has a prescribed linear form. The class of models, which we will refer to as canonical link regression models, constitutes arguably the most important subclass of generalized linear models and includes several of the most popular generalized linear models. In addition to the primary contribution of this study, we will revisit several other tests in the existing literature. The common feature among the proposed test, as well as the existing tests, is that they are all based on orthogonal series estimators and used to detect departures from a null model. Our proposal for a new lack-of-fit test is inspired by the recent contribution of Hart and is based on a Laplace approximation to the posterior probability of the null hypothesis. Despite having a Bayesian construction, the resulting statistic is implemented in a frequentist fashion. The formulation of the statistic is based on characterizing departures from the predictor function in terms of Fourier coefficients, and subsequent testing that all of these coefficients are 0. The resulting test statistic can be characterized as a weighted sum of exponentiated squared Fourier coefficient estimators, whereas the weights depend on user-specified prior probabilities. The prior probabilities provide the investigator the flexibility to examine specific departures from the prescribed model. Alternatively, the use of noninformative priors produces a new omnibus lack-of-fit statistic. We present a thorough numerical study of the proposed test and the various existing orthogonal series-based tests in the context of the logistic regression model. Simulation studies demonstrate that the test statistics under consideration possess desirable power properties against alternatives that have been identified in the existing literature as being important.Item Tunable bound-states in continuum by optical frequency(2013-12) Boretz, Yingyue Li; Reichl, Linda E.; Petrosky, Tomio Y.We demonstrate the existence of tunable bound-states in continuum (BIC) in a 1-dimensional quantum wire with two impurities by an intense monochromatic radiation field. We found that there is a new type of BIC due to the Fano interference between two optical transition channels, in addition to the ordinary BIC due to a geometrical interference between electron wave functions emitted by impurities. In both cases the BIC can be achieved by tuning the frequency of the radiation field.