Bayesian joint modeling of longitudinal and survival data

The joint modeling of longitudinal and survival data is a new approach to many applications such as HIV, cancer vaccine trials and quality of life studies. There are recent developments of the methodologies with respect to each of the components of the joint model as well as statistical processes that link them together. Among these, second order polynomial random effect models and linear mixed effects models are the most commonly used for the longitudinal trajectory function. In this study, we first relax the parametric constraints for polynomial random effect models by using Dirichlet process priors, then three longitudinal markers rather than only one marker are considered in one joint model. Second, we use a linear mixed effect model for the longitudinal process in a joint model analyzing the three markers. In this research these methods were applied to the Primary Biliary Cirrhosis sequential data, which were collected from a clinical trial of primary biliary cirrhosis (PBC) of the liver. This trial was conducted between 1974 and 1984 at the Mayo Clinic. The effects of three longitudinal markers (1) Total Serum Bilirubin, (2) Serum Albumin and (3) Serum Glutamic-Oxaloacetic transaminase (SGOT) on patients' survival were investigated. Proportion of treatment effect will also be studied using the proposed joint modeling approaches. ^ Based on the results, we conclude that the proposed modeling approaches yield better fit to the data and give less biased parameter estimates for these trajectory functions than previous methods. Model fit is also improved after considering three longitudinal markers instead of one marker only. The results from analysis of proportion of treatment effects from these joint models indicate same conclusion as that from the final model of Fleming and Harrington (1991), which is Bilirubin and Albumin together has stronger impact in predicting patients' survival and as a surrogate endpoints for treatment. ^

Assessment of the effect on statistical power of regression model misspecification by using techniques of mathematical statistics and simulation study

Objectives. This paper seeks to assess the effect on statistical power of regression model misspecification in a variety of situations. ^ Methods and results. The effect of misspecification in regression can be approximated by evaluating the correlation between the correct specification and the misspecification of the outcome variable (Harris 2010).In this paper, three misspecified models (linear, categorical and fractional polynomial) were considered. In the first section, the mathematical method of calculating the correlation between correct and misspecified models with simple mathematical forms was derived and demonstrated. In the second section, data from the National Health and Nutrition Examination Survey (NHANES 2007-2008) were used to examine such correlations. Our study shows that comparing to linear or categorical models, the fractional polynomial models, with the higher correlations, provided a better approximation of the true relationship, which was illustrated by LOESS regression. In the third section, we present the results of simulation studies that demonstrate overall misspecification in regression can produce marked decreases in power with small sample sizes. However, the categorical model had greatest power, ranging from 0.877 to 0.936 depending on sample size and outcome variable used. The power of fractional polynomial model was close to that of linear model, which ranged from 0.69 to 0.83, and appeared to be affected by the increased degrees of freedom of this model.^ Conclusion. Correlations between alternative model specifications can be used to provide a good approximation of the effect on statistical power of misspecification when the sample size is large. When model specifications have known simple mathematical forms, such correlations can be calculated mathematically. Actual public health data from NHANES 2007-2008 were used as examples to demonstrate the situations with unknown or complex correct model specification. Simulation of power for misspecified models confirmed the results based on correlation methods but also illustrated the effect of model degrees of freedom on power.^

B, d11B, K and d18O data for altered oceanic crust at DSDP Sites 417/418

Samples of drilled oceanic crust, from DSDP Holes 417A, 417D and 418A and ODP Hole 735B, and oceanic crust from the Oman and Cyprus ophiolites, were analyzed for B contents and d11B. Composite samples from DSDP Holes 417A, 417D and 418A were used to represent the upper 550 m of altered oceanic crustal Layer 2A. Whole-rock samples from the Troodos ophiolite, Cyprus, and the Oman ophiolite were selected to represent crustal Layer 2B dikes. Composite samples from ODP Hole 735B were used to represent crustal Layer 3. The B content of the DSDP composites ranges from 7.2 ppm to 104 ppm and correlates with both d1818O and K, showing that it is a good indicator of the extent of low temperature alteration. The d11B of the DSDP composites varies between -2.5? and 5.4?. The B content of the samples from the Troodos ophiolite ranges from 2.4 ppm to 8.1 ppm; d11B varies from -0.9? to 7.8?. The B content of the Oman ophiolite samples ranges from 5.0 ppm to 11.1 ppm; d11B varies from -1.6? to 16.9?. The B content of the samples from ODP Hole 735B ranges from 1.1 ppm to 7.1 ppm; d11B varies from -4.3? to 24.9?. The general pattern displayed by these samples is one of greatest (and most variable) B enrichment at the top of the crust and least enrichment at the bottom of the section. All of these samples are enriched compared to unaltered MORB, which is believed to have a B content of approximately 0.5 ppm. The d11B values of deeper samples, from Layers 2B and 3, are more variable and generally higher than those from Layer 2A. Boron contents and d11B are not correlated. The data from the DSDP Site 417/418 composites indicate that the d11B of fluid circulating in the upper crust changes only slightly during alteration, increasing by an average of 5.1? with an accompanying decrease in B concentration of 7%. Low temperature alteration appears to be a water-dominated process resulting in minor modification of circulating seawater. A minimum water-rock ratio of 400 is calculated for these samples, implying a minimum low-temperature seawater flux through the upper oceanic crust of 3.4?10**14 l/y. The average B content of altered oceanic crust, as represented by these samples, is 5.2+/-1.7 ppm and the average d11B is 3.4+/-1.1?. This average isotopic composition is measurably different from the apparent average of oceanic sediments, supporting the idea that d11B could be useful for identifying the source(s) of B in island arcs.

Decade Wind and Wave Data for NW Iberia, Cabo Sillerio

Seafloor sediment mobilization on the inner Northwest Iberian continental shelf is caused largely by ocean surface waves. The temporal and spatial variability in the wave height, wave period, and wave direction has a profound effect on local sediment mobilization, leading to distinct sediment mobilization scenarios. Six grain-size specific sediment mobilization scenarios, representing seasonal average and storm conditions, were simulated with a physics-based numerical model. Model inputs included meteorological and oceanographic data in conjunction with seafloor grain-size and the shelf bathymetric data. The results show distinct seasonal variations, most importantly in wave height, leading to sediment mobilization, specifically on the inner shelf shallower than 30 m water depth where up to 49% of the shelf area is mobilized. Medium to severe storm events are modeled to mobilize up to 89% of the shelf area above 150 m water depth. The frequency of each of these seasonal and storm-related sediment mobilization scenarios is addressed using a decade of meteorological and oceanographic data. The temporal and spatial patterns of the modeled sediment mobilization scenarios are discussed in the context of existing geological and environmental processes and conditions to assist scientific, industrial and environmental efforts that are directly affected by sediment mobilization. Examples, where sediment mobilization plays a vital role, include seafloor nutrient advection, recurrent arrival of oil from oil-spill-laden seafloor sediment, and bottom trawling impacts.

Organic- and carbonate-carbon values and extract data for DSDP Holes 63-467 and 63-471

Quantity, type, and maturity of the organic matter of middle Miocene to Quaternary sediments from the eastern North Pacific (Deep Sea Drilling Project Leg 63) were determined. Hydrocarbons and fatty acids in lipid extracts were analyzed by capillary column gas chromatography and combined gas chromatography/mass spectrometry. Kerogens were investigated by Rock-Eval pyrolysis and microscopy, and vitrinite reflectance values were determined. At Site 467, in the San Miguel Gap of the outer California Continental Borderland, organic carbon contents range from 1.46% to 5.40%. Normalized to organic carbon, total extracts increase from about 10 to 36 mg/g Corg with depth. The organic matter is a mixture of both marine and terrestrial origin, with the marine organic matter representing a high proportion in some of the samples. Steroid hydrocarbons - sterenes and steradienes in the upper part of the section and steranes in the deepest sample - are the most abundant compounds in the nonaromatic hydrocarbon fractions. Perylene, alkylated thiophenes, and aromatic steroid hydrocarbons dominate in the aromatic hydrocarbon fractions of the shallower samples; increasing maturation is indicated by a more petroleumlike aromatic hydrocarbon distribution. Microscopy revealed a high amount of liptinitic organic matter and confirmed the maturation trend as observed from analysis of the extracts. The vitrinite reflectance may be extrapolated to a bottom-hole value of nearly 0.5% Ro. The liquid hydrocarbon potential of the sediments at higher maturity levels is rated to be good to excellent. At Site 471, off Baja California, organic carbon values are between 0.70% and 1.12%. Extract values increase with depth, as at Site 467. The investigation of the soluble and insoluble organic matter, despite some compositional similarities, consistently revealed a more terrigenous influx compared with Site 467. Thus the potential for liquid hydrocarbon generation is lower, the organic matter being more gas-prone. The deepest sample analyzed indicates the onset of hydrocarbon generation. At this site, frequent sand intercalations offer pathways for migration and possibly reservoir formation.

Concentrations and accumulation rates of platinum-group elements and rhenium in Pacific and Atlantic sediments, DSDP and ODP data

The nature of Re-platinum-group element (PGE; Pt, Pd, Ir, Os, Ru) transport in the marine environment was investigated by means of marine sediments at and across the Cretaceous-Tertiary boundary (KTB) at two hemipelagic sites in Europe and two pelagic sites in the North and South Pacific. A traverse across the KTB in the South Pacific pelagic clay core found elevated levels of Re, Pt, Ir, Os, and Ru, each of which is approximately symmetrically distributed over a distance of ~1.8 m across the KTB. The Re-PGE abundance patterns are fractionated from chondritic relative abundances: Ru, Pt, Pd, and Re contents are slightly subchondritic relative to Ir, and Os is depleted by ~95% relative to chondritic Ir proportions. A similar depletion in Os (~90%) was found in a sample of the pelagic KTB in the North Pacific, but it is enriched in Ru, Pt, Pd, and Re relative to Ir. The two hemipelagic KTB clays have near-chondritic abundance patterns. The ~1.8-m-wide Re-PGE peak in the pelagic South Pacific section cannot be reconciled with the fallout of a single impactor, indicating that postdepositional redistribution has occurred. The elemental profiles appear to fit diffusion profiles, although bioturbation could have also played a role. If diffusion had occurred over ~65 Ma, the effective diffusivities are ~10**?13 cm**2/s, much smaller than that of soluble cations in pore waters (~10**?6 cm**2/s). The coupling of Re and the PGEs during redistribution indicates that postdepositional processes did not significantly fractionate their relative abundances. If redistribution was caused by diffusion, then the effective diffusivities are the same. Fractionation of Os from Ir during the KTB interval must therefore have occurred during aqueous transport in the marine environment. Distinctly subchondritic Os/Ir ratios throughout the Cenozoic in the South Pacific core further suggest that fractionation of Os from Ir in the marine environment is a general process throughout geologic time because most of the inputs of Os and Ir into the ocean have Os/Ir ratios >/=1. Mass balance calculations show that Os and Re burial fluxes in pelagic sediments account for only a small fraction of the riverine Os (<10%) and Re (<0.1%) inputs into the oceans. In contrast, burial of Ir in pelagic sediments is similar to the riverine Ir input, indicating that pelagic sediments are a much larger repository for Ir than for Os and Re. If all of the missing Os and Re is assumed to reside in anoxic sediments in oceanic margins, the calculated burial fluxes in anoxic sediments are similar to observed burial fluxes. However, putting all of the missing Os and Re into estuarine sediments would require high concentrations to balance the riverine input and would also fail to explain the depletion of Os at pelagic KTB sites, where at most ~25% of the K-T impactor's Os could have passed through estuaries. If Os is preferentially sequestered in anoxic marine environments, it follows that the Os/Ir ratio of pelagic sediments should be sensitive to changes in the rates of anoxic sediment deposition. There is thus a clear fractionation of Os and Re from Ir in precipitation out of sea water in pelagic sections. Accordingly, it is inferred here that Re and Os are removed from sea water in anoxic marine depositional regimes.

Water temperature and salinity data from southern elephant seals from Marion Island

Over the last decade, several hundred seals have been equipped with conductivity-temperature-depth sensors in the Southern Ocean for both biological and physical oceanographic studies. A calibrated collection of seal-derived hydrographic data is now available, consisting of more than 165,000 profiles. The value of these hydrographic data within the existing Southern Ocean observing system is demonstrated herein by conducting two state estimation experiments, differing only in the use or not of seal data to constrain the system. Including seal-derived data substantially modifies the estimated surface mixedlayer properties and circulation patterns within and south of the Antarctic Circumpolar Current. Agreement with independent satellite observations of sea ice concentration is improved, especially along the East Antarctic shelf. Instrumented animals efficiently reduce a critical observational gap, and their contribution to monitoring polar climate variability will continue to grow as data accuracy and spatial coverage increase.