The UNIVERSITY OF BRITISH COLUMBIA DEPARTMENT OF STATISTICS TECHNICAL REPORT #250 BAYESIAN EMPIRICAL ORTHOGONAL FUNCTIONS BY
نویسندگان
چکیده
This report presents a Bayesian version of empirical orthogonal functions (EOFs) and thereby overcomes a difficulty with the classical version of these functions when an environmental process is autocorrelated as most are. The approach partitions the spatial variation into long and short scale variation, the latter including measurement errors. The most general version of our results incorporates the generalized Wishart prior distribution for the unknown covariance matrices and thereby gains considerable flexibility. In particular, the method can contend with situations in which the data exhibit a monotone pattern of missingness. The report includes a simulation study that demonstrates how the proposed method better characterizes spatial pattern by removing the deleterious effect of autocorrelation.
منابع مشابه
THE UNIVERSITY OF BRITISH COLUMBIA DEPARTMENT OF STATISTICS TECHNICAL REPORT #252 Bayesian inference of gene-environment interaction from incomplete data: What happens when information on environment is disjoint from data on gene and disease?
متن کامل
THE UNIVERSITY OF BRITISH COLUMBIA DEPARTMENT OF STATISTICS TECHNICAL REPORT # 243 Asymptotics of Bayesian Median Loss Estimation By Chi
Here we define estimators based on minimizing the median of a loss function in the Bayesian context. For these estimators, we establish √ n-consistency, asymptotic normality and efficiency. We contrast the asymptotic behavior of these medloss estimators with the n asymptotic behavior of the least median of squares (LMS) estimators and the least trimmed squares (LTS) estimators which are √ n-con...
متن کاملUniversity of British Columbia Department of Statistics Technical Report # 256 April 2010 “ Selecting a binary Markov model for a precipitation process
This paper uses rth-order categorical Markov chains to model the probability of precipitation. Several stationary and non-stationary high order Markov models are proposed and compared using BIC. The number of parameters increases exponentially by adding the Markov order. Several classes of high-order Markov models are proposed which their increase of number of parameters are modest. The theory ...
متن کاملEffect of Co-existing Heavy Metals and Natural Organic Matter on Sorption/Desorption of Polycyclic Aromatic Hydrocarbons in Soil: A Review
Polycyclic aromatic hydrocarbons (PAHs), abundant in mixed contaminant sites, often coexist with heavy metals. The fate and remediation of PAHs depend heavily on the sorption and desorption behavior of these contaminants. The sorption behavior can in turn be highly affected by certain soil components and properties, such as soil organic matter (SOM) and the presence of heavy metals. Through rev...
متن کامل