Markov chain Monte Carlo methods for high dimensional inversion in remote sensing

Summary.  We discuss the inversion of the gas profiles (ozone, NO₃, NO₂, aerosols and neutral density) in the upper atmosphere from the spectral occultation measurements. The data are produced by the 'Global ozone monitoring of occultation of stars' instrument on board the Envisat satellite that was launched in March 2002. The instrument measures the attenuation of light spectra at various horizontal paths from about 100 km down to 10–20 km. The new feature is that these data allow the inversion of the gas concentration height profiles. A short introduction is given to the present operational data management procedure with examples of the first real data inversion. Several solution options for a more comprehensive statistical inversion are presented. A direct inversion leads to a non-linear model with hundreds of parameters to be estimated. The problem is solved with an adaptive single-step Markov chain Monte Carlo algorithm. Another approach is to divide the problem into several non-linear smaller dimensional problems, to run parallel adaptive Markov chain Monte Carlo chains for them and to solve the gas profiles in repetitive linear steps. The effect of grid size is discussed, and we present how the prior regularization takes the grid size into account in a way that effectively leads to a grid-independent inversion.     

A Bayesian method for classification and discrimination

We discuss Bayesian analyses of traditional normal-mixture models for classification and discrimination. The development involves application of an iterative resampling approach to Monte Carlo inference, commonly called Gibbs sampling, and demonstrates routine application. We stress the benefits of exact analyses over traditional classification and discrimination techniques, including the ease with which such analyses may be performed in a quite general setting, with possibly several normal-mixture components having different covariance matrices, the computation of exact posterior classification probabilities for observed data and for future cases to be classified, and posterior distributions for these probabilities that allow for assessment of second-level uncertainties in classification.     

State of the unions: Assessment by elite American labor leaders

One hundred and sixty-four elite union leaders in the United States completed a survey to assess the condition of the American labor movement and the factors affecting it. The respondents included high-level international union and state federation officers, central labor presidents, and 58 shop stewards. This study compares the responses of these labor officials and finds that they have similar, negative assessments of the condition of the American labor movement. The study also finds that these leaders agree that the most important factors affecting the labor movement are collective bargaining rights, union leadership, union member solidarity, and the NLRB.     

Assessment of influence in non-linear measurement error models

We consider the assessment of outliers and influential observation in non-linear measurement error models. Residuals, leverage measures and case-deletiondiagonostics are examined. The method of local influence is also applied to the models. In particular, the perturbation of measurement error variances has been found useful in assessing the adequancy of the model assumptions. A numerical example is given to illustrate the application of the diagonostics.     

A puzzlement: health-related organisations in the Nonprofit Almanac

Conclusion Data systems created with one purpose inevitably have shortcomings when used for other purposes, so it is not surprising that a database compiled from informational tax returns has serious flaws when used to compile a statistical picture of the non-profit sector. The two-digit NTEE level of aggregation that is published in theAlmanac conceals many problems that appear when the IRS-based data are broken down to the three-digit level.Identification of the problems in the NTEE and in the use of the IRS data is a necessary step toward their correction. At this point, it is clear that the IRS-based numbers in theAlmanac should be regarded, at best, as rough approximations and as a starting point for additional work. My assessment of theAlmanac suggests that, at least as far as health care is concerned, the NTEE needs revision. Correcting the problems with the IRS data will be much, much more difficult.I am grateful for assistance from Steve Noga, Chris Toppe and Virginia Hodgkinson at Independent Sector.