A Comparison of Bayesian and Sampling Theory Inferences in a Probit Model

HE PROBIT MODEL IS A POPULAR DEVICE for explaining binary choice decisions in econometrics. It has been used to describe choices such as labor force participation, travel mode, home ownership, and type of education. These and many more examples can be found in papers by Amemiya (1981) and Maddala (1983). Given the contribution of economics towards explaining such choices, and given the nature of data that are collected, prior information on the relationship between a choice probability and several explanatory variables frequently exists. Bayesian inference is a convenient vehicle for including such prior information. Given the increasing popularity of Bayesian inference it is useful to ask whether inferences from a probit model are sensitive to a choice between Bayesian and sampling theory techniques. Of interest is the sensitivity of inference on coefficients, probabilities, and elasticities. We consider these issues in a model designed to explain choice between fixed and variable interest rate mortgages. Two Bayesian priors are employed: a uniform prior on the coefficients, designed to be noninformative for the coefficients, and an inequality restricted prior on the signs of the coefficients. We often know, a priori, whether increasing the value of a particular explanatory variable will have a positive or negative effect on a choice probability. This knowledge can be captured by using a prior probability density function (pdf) that is truncated to be positive or negative. Thus, three sets of results are compared:those from maximum likelihood (ML) estimation, those from Bayesian estimation with an unrestricted uniform prior on the coefficients, and those from Bayesian estimation with a uniform prior truncated to accommodate inequality restrictions on the coefficients.

Phylogeny of the filamentous bacterium 'Nostocoida limicola' III from activated sludge

Five strains of the filamentous bacterium 'Nostocoida limicola' III were successfully isolated into pure culture from samples of activated sludge biomass from five plants in Australia. 16S rRNA gene sequence analyses showed that all isolates were members of the Planctomycetales, most closely related to Isosphaera pallida, but they differed phenotypically from this species in that they did not glide and were not thermotolerant. The ultrastructure of these 'N. limicola' III isolates was also consistent with them being Planctomycetales, in that they possessed complex intracellular membrane systems compartmentalizing the cells. However, the arrangements of these intracellular membranes differed between isolates. These data confirm that 'N. limicola' III is phylogenetically unrelated to both 'N. limicola' I and 'N. limicola' II, activated sludge filamentous bacteria which share morphological features in common with 'N. limicola' III and which have been presumed historically to be the same or very similar bacteria.

A total-evidence phylogeny of ticks provides insights into the evolution of life cycles and biogeography

We inferred the phylogeny of 33 species of ticks from the subfamilies Rhipicephalinae and Hyalomminae from analyses of nuclear and mitochondrial DNA and morphology. We used nucleotide sequences from 12S rRNA, cytochrome c oxidase I, internal transcribed spacer 2 of the nuclear rRNA, and 18S rRNA. Nucleotide sequences and morphology were analyzed separately and together in a total-evidence analysis. Analyses of the five partitions together (3303 characters) gave the best-resolved and the best-supported hypothesis so far for the phylogeny of ticks in the Rhipicephalinae and Hyalomminae, despite the fact that some partitions did not have data for some taxa. However, most of the hidden conflict (lower support in the total-evidence analyses compared to that in the individual analyses) was found in those partitions that had taxa without data. The partitions with complete taxonomic sampling had more hidden support (higher support in the total-evidence analyses compared to that in the separate-partition analyses) than hidden conflict. Mapping of geographic origins of ticks onto our phylogeny indicates an African origin for the Rhipicephalinae sensu lato (i.e., including Hyalomma spp.), the Rhipicephalus-Boophilus lineage, the Dermacentor-Anocentor lineage, and the Rhipicephalus-Booophilus-Nosomma-Hyalomma-Rhipicentor lineage. The Nosomma-Hyalomma lineage appears to have evolved in Asia. Our total-evidence phylogeny indicates that (i) the genus Rhipicephalus is paraphyletic with respect to the genus Boophilus, (ii) the genus Dermacentor is paraphyletic with respect to the genus Anocentor, and (iii) some subgenera of the genera Hyalomma and Rhipicephalus are paraphyletic with respect to other subgenera in these genera. Study of the Rhipicephalinae and Hyalomminae over the last 7 years has shown that analyses of individual datasets (e.g., one gene or morphology) seldom resolve many phylogenetic relationships, but analyses of more than one dataset can generate well-resolved phylogenies for these ticks. (C) 2001 Academic Press.

A phylogeny for the genus Isoodon and a range extension for I-obesulus peninsulae based on mtDNA control region and morphology

Short-nosed bandicoots, Isoodon, have undergone marked range contractions since European colonisation of Australia and are currently divided into many subspecies, the validity of which is debated. Discriminant function analysis of morphology and a phylogeny of Isoodon based on mtDNA control region sequences indicate a clear split between two of the three recognised species, I. macrourus and I. obesulus/auratus. However, while all previously recognised taxa within the I. obesulus/auratus group are morphologically distinct, I. auratus and I. obesulus are not phylogenetically distinct for mtDNA. The genetic divergence between I. obesulus and I. auratus (2.6%) is similar to that found among geographic isolates of the former (I. o. obesulus and I. o. peninsulae: 2.7%). Further, the divergence between geographically close populations of two different species (I. o. obesulus from Western Australia and I. a. barrowensis: 1.2%) is smaller than that among subspecies within I. auratus (I. a. barrowensis and I. auratus from northern Western Australia: 1.7%). A newly discovered population of Isoodon in the Lamb Range, far north Queensland, sympatric with a population of I. m. torosus, is shown to represent a range extension of I. o. peninsulae (350 km). It seems plausible that what is currently considered as two species, I. obesulus and I. auratus, was once one continuous species now represented by isolated populations that have diverged morphologically as a consequence of adaptation to the diverse environments that occur throughout their range. The taxonomy of these populations is discussed in relation to their morphological distinctiveness and genetic similarity.

Bayesian feedback versus Markovian feedback in a two-level atom

We compare two different approaches to the control of the dynamics of a continuously monitored open quantum system. The first is Markovian feedback, as introduced in quantum optics by Wiseman and Milburn [Phys. Rev. Lett. 70, 548 (1993)]. The second is feedback based on an estimate of the system state, developed recently by Doherty and Jacobs [Phys. Rev. A 60, 2700 (1999)]. Here we choose to call it, for brevity, Bayesian feedback. For systems with nonlinear dynamics, we expect these two methods of feedback control to give markedly different results. The simplest possible nonlinear system is a driven and damped two-level atom, so we choose this as our model system. The monitoring is taken to be homodyne detection of the atomic fluorescence, and the control is by modulating the driving. The aim of the feedback in both cases is to stabilize the internal state of the atom as close as possible to an arbitrarily chosen pure state, in the presence of inefficient detection and other forms of decoherence. Our results (obtained without recourse to stochastic simulations) prove that Bayesian feedback is never inferior, and is usually superior, to Markovian feedback. However, it would be far more difficult to implement than Markovian feedback and it loses its superiority when obvious simplifying approximations are made. It is thus not clear which form of feedback would be better in the face of inevitable experimental imperfections.

A molecular phylogeny of rainbow skinks (Scincidae : Carlia): taxonomic and biogeographic implications

The phylogenetic relationships amongst 29 species of Carlia and Lygisaurus were estimated using a 726-base-pair segment of the protein-coding mitochondrial ND4 gene. Results do not support the recent resurrection of the genus Lygisaurus. Although most Lygisaurus species formed a single clade, this clade is nested within Carlia and includes Carlia parrhasius. Due to this new molecular evidence, and the paucity of diagnostic morphological characters separating the genera, Lygisaurus de Vis 1884 is re-synonymised with Carlia Gray 1845. Our analysis is also inconsistent with a previous suggestion that Lygisaurus timlowi should be removed to Menetia, a genus that is distantly related relative to outgroups used here. Intraspecific variation in Carlia is, in several instances, greater than interspecific distance. The most strikingly divergent lineages are found within C. rubrigularis, which appears to be paraphyletic, with southern populations more closely related to C. rhomboidalis than to northern populations of C. rubrigularis. The two C. rubrigularis-C. rhomboidalis lineages form part of a major polytomy at an intermediate level of divergence. Lack of resolution at this level, however, does not appear to be due to saturation or loss of phylogenetic signal. Rather, the polytomy probably reflects a period of relatively rapid diversification that occurred sometime during the Miocene.

Argyrodes: phylogeny, sociality and interspecific interactions - a report on the Argyrodes symposium, Badplaas 2001

Argyrodes Simon 1864 is a large, cosmopolitan theridiid genus whose members exhibit a wide range of foraging techniques which usually involve exploiting other spiders, either by using their webs, stealing their food, or preying on them directly. We held a symposium on this genus at the 15th International Congress of Arachnology, Badplaas, South Africa in order to obtain a clearer perspective on the relationship between the phylogeny of the genus and the different foraging techniques. We concluded that Argyrodes forms a monophyletic group within the Theridiidae, and that there are clear monophyletic clades within the genus (already identified as species groups) that appear to share behavioral characteristics. We found no clear indication that foraging behaviors such as kleptoparasitism (stealing food) evolved from araneophagy (eating spiders) or vice versa. However, it appears that species that specialize in either kleptoparasitism or araneophagy use additional techniques in comparison to species that readily use both foraging modes. During our examination of Argyrodes/host interactions we noted the importance of Nephila species as hosts of Argyrodes species around the world and the impact of Argyrodes on Nephila. We also noted the fluid nature of the relationship between Argyrodes and the spiders with which they interact. For example, an Argyrodes/host relationship can change to an Argyrodes/prey relationship, and the type of kleptoparasitic behavior employed by an Argyrodes can change when it changes host species. The importance of eating silk was also noted and identified as an area for further research. We concluded that more work involving international collaboration is needed to fully understand the phylogeny of the genus and the relationships between the different types of foraging behaviors.