Inference in credal networks using multilinear programming

A credal network is a graphical tool for representation and manipulation of uncertainty, where probability values may be imprecise or indeterminate. A credal network associates a directed acyclic graph with a collection of sets of probability measures; in this context, inference is the computation of tight lower and upper bounds for conditional probabilities. In this paper we present new algorithms for inference in credal networks based on multilinear programming techniques. Experiments indicate that these new algorithms have better performance than existing ones, in the sense that they can produce more accurate results in larger networks.

Eliciting and estimating valid subjective probabilities: An experimental investigation of the Exchangeability Method.

Using a laboratory experiment, we investigate whether incentive compatibility affects subjective probabilities elicited via the exchangeability method (EM), an elicitation technique consisting of several chained questions. We hypothesize that subjects who are aware of the chaining strategically behave and provide invalid subjective probabilities, while subjects who are not aware of the chaining state their real beliefs and provide valid subjective probabilities. The validity of subjective probabilities is investigated using de Finetti's notion of coherence, under which probability estimates are valid if and only if they obey all axioms of probability theory.
Four experimental treatments are designed and implemented. Subjects are divided into two initial treatment groups: in the first, they are provided with real monetary incentives, and in the second, they are not. Each group is further sub-divided into two treatment groups, in the first, the chained structure of the experimental design is made clear to the subjects, while, in the second, the chained structure is hidden by randomizing the elicitation questions.
Our results suggest that subjects provided with monetary incentives and randomized questions provide valid subjective probabilities because they are not aware of the chaining which undermines the incentive compatibility of the exchangeability method.

Estimating the masses of extra-solar planets

All extra-solar planet masses that have been derived spectroscopically are lower limits since the inclination of the orbit to our line-of-sight is unknown except for transiting systems. In theory, however, it is possible to determine the inclination angle, i, between the rotation axis of a star and an observer's line-of-sight from measurements of the projected equatorial velocity (v sin i), the stellar rotation period (P(rot)) and the stellar radius (R(*)). For stars which host planetary systems this allows the removal of the sin i dependency of extra-solar planet masses derived from spectroscopic observations under the assumption that the planetary orbits lie perpendicular to the stellar rotation axis.
We have carried out an extensive literature search and present a catalogue of v sin i, P(rot) and R(*) estimates for stars hosting extra-solar planets. In addition, we have used Hipparcos parallaxes and the Barnes-Evans relationship to further supplement the R(*) estimates obtained from the literature. Using this catalogue, we have obtained sin i estimates using a Markov-chain Monte Carlo analysis. This technique allows proper 1 Sigma two-tailed confidence limits to be placed on the derived sin i's along with the transit probability for each planet to be determined.
While we find that a small proportion of systems yield sin i's significantly greater than 1, most likely due to poor P(rot) estimations, the large majority are acceptable. We are further encouraged by the cases where we have data on transiting systems, as the technique indicates inclinations of similar to 90 degrees and high transit probabilities. In total, we are able to estimate the true masses of 133 extra-solar planets. Of these 133 extra-solar planets, only six have revised masses that place them above the 13M(J) deuterium burning limit; four of those six extra-solar planet candidates were already suspected to lie above the deuterium burning limit before correcting their masses for the sin i dependency. Our work reveals a population of high-mass extra-solar planets with low eccentricities, and we speculate that these extra-solar planets may represent the signature of different planetary formation mechanisms at work. Finally, we discuss future observations that should improve the robustness of this technique.

Energy levels and transition probabilities for nitrogen-like Fe xx

Energies of the 700 lowest levels in Fe XX have been obtained using the multiconfiguration Dirac-Fock method. Configuration interaction method on the basis set of transformed radial orbitals with variable parameters taking into account relativistic corrections in the Breit-Pauli approximation was used to crosscheck our presented results. Transition probabilities, oscillator and line strengths are presented for electric dipole (E1), electric quadrupole (E2) and magnetic dipole (M1) transitions among these levels. The total radiative transition probabilities from each level are also provided. Results are compared with data compiled by NIST and with other theoretical work.

Relativistic allowed and forbidden transition probabilities for fluorine-like Fe XVIII

Energy levels and the corresponding transition probabilities for allowed and forbidden transitions among the levels of the ground configuration and first 23 excited configurations of fluorine-like Fe XVIII have been calculated using the multiconfigurational Dirac-Fock GRASP code. A total of 379 lowest bound levels of Fe XVIII is presented, and the energy levels are identified in spectroscopic notations. Transition probabilities, oscillator strengths and line strengths for electric dipole (E1), electric quadrupole (E2) and magnetic dipole (M1) transitions among these 379 levels are also presented. The calculated energy levels and transition probabilities are compared with experimental data.