Fatores gravimétricos no Brasil para um modelo anelástico de Terra

Geodetic observations are affected by the disturbing potential of the luni-solar tide. Among those observations, the value of g obtained from gravimetric survey needs correction by the gravimetric factor. This correction is derived from the Numbers of Love, which depend on the adopted model of Earth. Because of this, it is necessary to update the correction since the gravimetric factor widely used in Brazil as delta = 1.20 does not consider local rheological variations and they are latitude dependent. A discrepancy of about 1% between the observed tidal gravimetric factors d of the ""Trans World Tidal Gravity Profiles"" (TWTGP), related to Brussels fundamental station, and those obtained by recent observations reported by Freitas and Ducarme ( 1991). Experiments based on inertial force effects also reveal a variation of about 0.5% in the observed d. A same order of magnitude difference is obtained for an anelastic Earth model when compared with a viscous-elastic model and even when different frequencies of tidal perturbations are considered. In this paper regression models are presented for gravimetric factors for the lunar components O(1) and M(2) in Brazil. These models were obtained from observations performed at stations belonging to the Brazilian segment of the TWTGP.

On estimation and influence diagnostics for zero-inflated negative binomial regression models

The zero-inflated negative binomial model is used to account for overdispersion detected in data that are initially analyzed under the zero-Inflated Poisson model A frequentist analysis a jackknife estimator and a non-parametric bootstrap for parameter estimation of zero-inflated negative binomial regression models are considered In addition an EM-type algorithm is developed for performing maximum likelihood estimation Then the appropriate matrices for assessing local influence on the parameter estimates under different perturbation schemes and some ways to perform global influence analysis are derived In order to study departures from the error assumption as well as the presence of outliers residual analysis based on the standardized Pearson residuals is discussed The relevance of the approach is illustrated with a real data set where It is shown that zero-inflated negative binomial regression models seems to fit the data better than the Poisson counterpart (C) 2010 Elsevier B V All rights reserved