THE LOG-BURR XII REGRESSION MODEL FOR GROUPED SURVIVAL DATA

The log-Burr XII regression model for grouped survival data is evaluated in the presence of many ties. The methodology for grouped survival data is based on life tables, where the times are grouped in k intervals, and we fit discrete lifetime regression models to the data. The model parameters are estimated by maximum likelihood and jackknife methods. To detect influential observations in the proposed model, diagnostic measures based on case deletion, so-called global influence, and influence measures based on small perturbations in the data or in the model, referred to as local influence, are used. In addition to these measures, the total local influence and influential estimates are also used. We conduct Monte Carlo simulation studies to assess the finite sample behavior of the maximum likelihood estimators of the proposed model for grouped survival. A real data set is analyzed using a regression model for grouped data.

Radiative transfer analyses of Titan's tropical atmosphere

Titan's optical and near-IR spectra result primarily from the scattering of sunlight by haze and its absorption by methane. With a column abundance of 92 km amagat (11 times that of Earth), Titan's atmosphere is optically thick and only similar to 10% of the incident solar radiation reaches the surface, compared to 57% on Earth. Such a formidable atmosphere obstructs investigations of the moon's lower troposphere and surface, which are highly sensitive to the radiative transfer treatment of methane absorption and haze scattering. The absorption and scattering characteristics of Titan's atmosphere have been constrained by the Huygens Probe Descent Imager/Spectral Radiometer (DISR) experiment for conditions at the probe landing site (Tomasko, M.G., Bezard, B., Doose, L., Engel, S., Karkoschka, E. 120084 Planet. Space Sci. 56, 624-247: Tomasko, M.G. et al. [2008b] Planet. Space Sci. 56, 669-707). Cassini's Visual and Infrared Mapping Spectrometer (VIMS) data indicate that the rest of the atmosphere (except for the polar regions) can be understood with small perturbations in the high haze structure determined at the landing site (Penteado, P.F., Griffith, CA., Tomasko, M.G., Engel, S., See, C., Doose, L, Baines, K.H., Brown, R.H., Buratti, B.J., Clark, R., Nicholson, P., Sotin, C. [2010]. Icarus 206, 352-365). However the in situ measurements were analyzed with a doubling and adding radiative transfer calculation that differs considerably from the discrete ordinates codes used to interpret remote data from Cassini and ground-based measurements. In addition, the calibration of the VIMS data with respect to the DISR data has not yet been tested. Here, VIMS data of the probe landing site are analyzed with the DISR radiative transfer method and the faster discrete ordinates radiative transfer calculation; both models are consistent (to within 0.3%) and reproduce the scattering and absorption characteristics derived from in situ measurements. Constraints on the atmospheric opacity at wavelengths outside those measured by DISR, that is from 1.6 to 5.0 mu m, are derived using clouds as diffuse reflectors in order to derive Titan's surface albedo to within a few percent error and cloud altitudes to within 5 km error. VIMS spectra of Titan at 2.6-3.2 mu m indicate not only spectral features due to CH4 and CH3D (Rannou, P., Cours, T., Le Mouelic, S., Rodriguez, S., Sotin, C., Drossart, P., Brown, R. [2010]. Icarus 208, 850-867), but also a fairly uniform absorption of unknown source, equivalent to the effects of a darkening of the haze to a single scattering albedo of 0.63 +/- 0.05. Titan's 4.8 mu m spectrum point to a haze optical depth of 0.2 at that wavelength. Cloud spectra at 2 mu m indicate that the far wings of the Voigt profile extend 460 cm(-1) from methane line centers. This paper releases the doubling and adding radiative transfer code developed by the DISR team, so that future studies of Titan's atmosphere and surface are consistent with the findings by the Huygens Probe. We derive the surface albedo at eight spectral regions of the 8 x 12 km(2) area surrounding the Huygens landing site. Within the 0.4-1.6 mu m spectral region our surface albedos match DISR measurements, indicating that DISR and VIMS measurements are consistently calibrated. These values together with albedos at longer 1.9-5.0 mu m wavelengths, not sampled by DISR, resemble a dark version of the spectrum of Ganymede's icy leading hemisphere. The eight surface albedos of the landing site are consistent with, but not deterministic of, exposed water ice with dark impurities. (C) 2011 Elsevier Inc. All rights reserved.

Influence diagnostics in Gaussian spatial linear models

Spatial linear models have been applied in numerous fields such as agriculture, geoscience and environmental sciences, among many others. Spatial dependence structure modelling, using a geostatistical approach, is an indispensable tool to estimate the parameters that define this structure. However, this estimation may be greatly affected by the presence of atypical observations in the sampled data. The purpose of this paper is to use diagnostic techniques to assess the sensitivity of the maximum-likelihood estimators, covariance functions and linear predictor to small perturbations in the data and/or the spatial linear model assumptions. The methodology is illustrated with two real data sets. The results allowed us to conclude that the presence of atypical values in the sample data have a strong influence on thematic maps, changing the spatial dependence structure.

STRUCTURE AND BIFURCATION OF PULLBACK ATTRACTORS IN A NON-AUTONOMOUS CHAFEE-INFANTE EQUATION

The Chafee-Infante equation is one of the canonical infinite-dimensional dynamical systems for which a complete description of the global attractor is available. In this paper we study the structure of the pullback attractor for a non-autonomous version of this equation, u(t) = u(xx) + lambda(xx) - lambda u beta(t)u(3), and investigate the bifurcations that this attractor undergoes as A is varied. We are able to describe these in some detail, despite the fact that our model is truly non-autonomous; i.e., we do not restrict to 'small perturbations' of the autonomous case.