"Development of sixteen polymorphic simple sequence repeat markers for Lycoris longituba from expressed sequence tags.". (with He et. al.) Lycoris longituba is a tulip like ornamental plant in China. Multiallelic markers were generated and analyzed for this species. Submitted.
"A Computational Approach to the Functional Clustering of Periodic Gene Expression Profiles.". (with Kim et. al.) This paper presents a new statistical model to the problem of clustering periodic patterns of gene expressions. Fourier series approximations are interwoven into a mixture model-based likelihood function where the number of components in the mixture are selected by an information criterion. The model reanalyzed data of 632 periodically expressed transcriptional genes in yeast originally analyzed in Spellman et. al. (Mol. Bio. Cell, 1998). Submitted.
"Computing genetic imprinting expressed by haplotypes". (with Cheng et. al.) This paper introduces a novel statistical model for detecting imprinted QTLs by haplotypes constructed with multiple SNPs. Applied to a human BMI dataset, a significantly imprinted effect was discovered at the haplotype level. Submitted.
"A Statistical Model for Testing the Pleiotropic Control of Phenotypic Plasticity for a Count Trait". (with Wu et. al.) This paper incorporates a multivariate Poisson distribution into the mixture model framework for QTL mapping of phenotypic plasticity. A hierarchical EM algorithm is intoduced to provide ML estimates of the multivariate Poisson distribution in the presence of a mixture model. Published in Genetics (2008).
"On Effect-Measure Modification: Relations Among Changes in the Relative Risk, Odds Ratio, and Risk Difference". (with Brumback) This paper delves into several relationships among the relative risk, odds ratio, and risk difference. Exact conditions are provided through rigerous inequality calculations. To appear in Statistics in Medicine.
"An Occupancy Problem Arising in Power Law Fitting". (with Abramson and Meyers) This paper focuses on power law fitting by least squares. The central theorem mathematically justifies a "delete-by-eye" approach for removing sparse cell counts. Interesting mathematics is abound including a Poissonization argument and calculations with the Reimann zeta function. Submitted.
"CDF and Survival Function Estimation with Infinite-Order Kernels". (with Politis) This paper uses infinite-order kernels to improve CDF and survival function estimation. The analysis uses sophisticated mathematics in the realm of generalized functions and Fourier transform theory. Novel asymptotic relative deficiency calculations are also included. Submitted.
"Density Estimation of Censored Data with Infinite-Order Kernels". (with Politis) This paper uses infinite-order kernels to improve nonparametric density estimation and hazard function estimation in censored data. Submitted to Biometrika.
"Multivariate Lag-Windows and Group Representations". This paper exhibits an interesting connection between the symmetries inherit in polyspectra and a discovered formulation of a familiar group representation of the symmetric group. Applications include symmetrizing multivariate kernels and generalizing the Subba-Rao Gabr optimal window. To appear in JMVA.
"Higher-Order Accurate Polyspectral Estimation with Flat-top Lag-Windows". (with Politis) This paper uses infinite-order kernels to improve polyspectra estimation. A generalized and more robust version of the bandwidth selection procedure proposed in Politis 2001 is also discussed. To appear in AISM.
"Quantifying the Remote Triggering Capabilities of Large Teleseismic Earthquakes Using Data From The ANZA Seismic Network Catalog". Figures. (with Kane, Kilb, and Martynov) This paper was a product of collaboration with researchers from the Scripps Institute of Oceanography at UCSD. Using a large database of teleseismic data from the ANZA region of Southern California, we proposed several statistical tests to identify triggering versus non-triggering of aftershocks. Published in JGR (2007).