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.

Globalisation and the technology gap: Regional and time evidence

This chapter examines the relationship between globalisation and technological progress. It computes an annual and country specific measure of technological gap, the technology ratio (TGR), using a recently proposed method known as metafrontiers. The TGR is measured as the distance from a group frontier to the global (or meta) frontier. The TGRs provide a measure to compare technological capability across countries. The ranking obtained from the metafrontiers method is first compared to other methods based on the direct measure of patents, science articles, schooling etc. The TGRs are then related to levels of trade openness and inbound and outbound foreign direct investment within regions and overtime in an effort to identify the relationship between technological gap and outward orientation.

Microbial and sensory changes throughout the ripening of Prato cheese made from milk with different levels of somatic cells

The objective of this research was to evaluate the effects of 2 levels of raw milk somatic cell count (SCC) on the composition of Prato cheese and on the microbiological and sensory changes of Prato cheese throughout ripening. Two groups of dairy cows were selected to obtain low-SCC (<200,000 cells/mL) and high-SCC (>700,000 cells/mL) milks, which were used to manufacture 2 vats of cheese. The pasteurized milk was evaluated according to the pH, total solids, fat, total protein, lactose, standard plate count, coliforms at 45 degrees C, and Salmonella spp. The cheese composition was evaluated 2 d after manufacture. Lactic acid bacteria, psychrotrophic bacteria, and yeast and mold counts were carried out after 3, 9, 16, 32, and 51 d of storage. Salmonella spp., Listeria monocytogenes, and coagulase-positive Staphylococcus counts were carried out after 3, 32, and 51 d of storage. A 2 x 5 factorial design with 4 replications was performed. Sensory evaluation of the cheeses from low- and high-SCC milks was carried out for overall acceptance by using a 9-point hedonic scale after 8, 22, 35, 50, and 63 d of storage. The somatic cell levels used did not affect the total protein and salt: moisture contents of the cheeses. The pH and moisture content were higher and the clotting time was longer for cheeses from high-SCC milk. Both cheeses presented the absence of Salmonella spp. and L. monocytogenes, and the coagulase-positive Staphylococcus count was below 1 x 10(2) cfu/g throughout the storage time. The lactic acid bacteria count decreased significantly during the storage time for the cheeses from both low- and high-SCC milks, but at a faster rate for the cheese from high-SCC milk. Cheeses from high-SCC milk presented lower psychrotrophic bacteria counts and higher yeast and mold counts than cheeses from low- SCC milk. Cheeses from low- SCC milk showed better overall acceptance by the consumers. The lower overall acceptance of the cheeses from high-SCC milk may be associated with texture and flavor defects, probably caused by the higher proteolysis of these cheeses.

Evaluation of lignin contents in tropical forages using different analytical methods and their correlations with degradation of insoluble fiber

We compared the lignin contents of tropical forages by different analytical methods and evaluated their correlations with parameters related to the degradation of neutral detergent fiber (NDF). The lignin content was evaluated by five methods: cellulose solubilization in sulfuric acid [Lignin (sa)], oxidation with potassium permanganate [Lignin (pm)], the Klason lignin method (KL), solubilization in acetyl bromide from acid detergent fiber (ABLadf) and solubilization in acetyl bromide from the cell wall (ABLcw). Samples from ten grasses and ten legumes were used. The lignin content values obtained by gravimetric methods were also corrected for protein contamination, and the corrected values were referred to as Lignin (sa)p, Lignin (pm)p and KLp. The indigestible fraction of NDF (iNDF), the discrete lag (LAG) and the fractional rate of degradation (kd) of NDF were estimated using an in vitro assay. Correcting for protein resulted in reductions (P < 0.05) in the lignin contents as measured by the Lignin (sa), Lignin (pm) and, especially, the KL methods. There was an interaction (P < 0.05) of analytical method and forage group for lignin content. In general, LKp method provided the higher (P < 0.05) lignin contents. The estimates of lignin content obtained by the Lignin (sa)p, Lignin (pm)p and LKp methods were associated (P > 0.05) with all of the NDF degradation parameters. However, the strongest correlation coefficients for all methods evaluated were obtained with Lignin (pm)p and KLp. The lignin content estimated by the ABLcw method did not correlate (P > 0.05) with any parameters of NDF degradation. There was a correlation (P < 0.05) between the lignin content estimated by the ABLadf method and iNDF content. Nonetheless, this correlation was weaker than those found with gravimetric methods. From these results, we concluded that the gravimetric methods produce residues that are contaminated by nitrogenous compounds. Adjustment for these contaminants is suggested, particularly for the KL method, to express lignin content with greater accuracy. The relationships between lignin content measurements and NDF degradation parameters can be better determined using KLp and Lignin (pm)p methods. (C) 2011 Elsevier B.V. All rights reserved.

Low genetic diversities of rabies virus populations within different hosts in Brazil

The low rates of nonsynonymous evolution observed in natural rabies virus (RABV) isolates are suggested to have arisen in association with the structural and functional constraints operating on the virus protein and the infection strategies employed by RABV within infected hosts to avoid strong selection by the immune response. In order to investigate the relationship between the genetic characteristics of RABV populations within hosts and the virus evolution, the present study examined the genetic heterogeneities of RABV populations within naturally infected dogs and foxes in Brazil, as well as those of bat RABV populations that were passaged once in suckling mice. Sequence analyses of complete RABV glycoprotein (G) genes showed that RABV populations within infected hosts were genetically highly homogeneous whether they were infected naturally or experimentally (nucleotide diversities of 0-0.95 x 10(-3)). In addition, amino acid mutations were randomly distributed over the entire region of the G protein, and the nonsynonymous/synonymous rate ratios (d(N)/d(S)) for the G protein gene were less than 1. These findings suggest that the low genetic diversities of RABV populations within hosts reflect the stabilizing selection operating on the virus, the infection strategies of the virus, and eventually, the evolutionary patterns of the virus. (C) 2009 Elsevier B.V. All rights reserved.

Effects of gonadotropin-releasing hormone at initiation of the 5-d timed artificial insemination (AI) program and timing of induction of ovulation relative to AI on ovarian dynamics and fertility of dairy heifers

Two experiments evaluated the effects of the first GnRH injection of the 5-d timed artificial insemination (AI) program on ovarian responses and pregnancy per AT (P/AI), and the effect of timing of the final GnRH to induce ovulation relative to AT on P/AI. In experiment 1, 605 Holstein heifers were synchronized for their second insemination and assigned randomly to receive GnRH on study d 0 (n = 298) or to remain as untreated controls (n = 307). Ovaries were scanned on study d 0 and 5. All heifers received a controlled internal drug-release (CIDR) insert containing progesterone on d 0, a single injection of PGF(2 alpha),, and removal of the CIDR on d 5, and GnRH concurrent with timed AT on d 8. Blood was analyzed for progesterone at AI. Pregnancy was diagnosed on d 32 and 60 after AI. Ovulation on study d 0 was greater for GnRH than control (35.4 vs. 10.6%). Presence of a new corpus luteum (CL) at PGF(2 alpha),, injection was greater for GnRH than for control (43.1 vs. 20.8%), although the proportion of heifers with a CL at PGF(2 alpha) did not differ between treatments and averaged 87.1%. Progesterone on the day of AT was greater for GaRH than control (0.50 +/- 0.07 vs. 0.28 +/- 0.07 ng/mL). The proportion of heifers at AI with progesterone <0.5 ng/mL was less for GURH than for control (73.8 vs. 88.2%). The proportion of heifers in estrus at AI did not differ between treatments and averaged 66.8%. Pregnancy per AI was not affected by treatment at d 32 or 60 (GnRH = 52.5 and 49.8% vs. control = 54.1 and 50.0%), and pregnancy loss averaged 6.0%. Responses to GnRH were not influenced by ovarian status on study d 0. In experiment 2, 1,295 heifers were synchronized for their first insemination and assigned randomly to receive a CIDR on d 0, PGF(2 alpha) and removal of the CIDR on d 5, and either GnRH 56 h after PGF(2 alpha) and AI 16 h later (OVS56, n = 644) or GnRH concurrent with AI 72 h after PGF(2 alpha) (COS72; n = 651). Estrus at AI was greater for COS72 than for OVS56 (61.4 vs. 47.5). Treatment did not affect P/AI on d 32 in heifers displaying signs of estrus at AI, but COS72 improved P/AI compared with OVS56 (55.0 vs. 47.6%) in those not in estrus at AI. Similarly, P/AI on d 60 did not differ between treatments for heifers displaying estrus, but COS72 improved P/AI compared with OVS56 (53.0 vs. 44.7%) in those not in estrus at AI. Administration of GnRH on the first day of the 5-d timed AI program resulted in low ovulation rate and no improvement in P/AI when heifers received a single PGF(2 alpha) injection 5 d later. Moreover, extending the proestrus by delaying the finAI GnRH from 56 to 72 h concurrent with AI benefited fertility of dairy heifers that did not display signs of estrus at insemination following the 5-d timed AI protocol.

Vitrification with Glutamine Improves Maturation Rate of Vitrified/Warmed Immature Bovine Oocytes

Contents The current study examined the protective effects of l-glutamine and cytochalasin B during vitrification of immature bovine oocytes. Oocyte vitrification solution (PBS supplemented with 10% FCS, 25% EG, 25% DMSO and 0.5 m trehalose) was the vitrification control. Treatments were the addition of 7 mu g/ml cytochalasin B, 80 mm glutamine or both cytochalasin and glutaminine for 30 s. After warming, oocytes were matured in vitro for 24 h, fixed and stained with Hoechst (33342) for nuclear maturation evaluation. l-glutamine improved the vitrified/warmed immature bovine oocytes viability (32.8%), increasing the nuclear maturation rates compared to other treatments and the no treatment vitrified control (17.4%). There was, however, no effect of cytochalasin B on in vitro maturation (14.4%).

Maned wolf survival rate in central Brazil

Although many carnivores are of conservation concern, most are poorly studied. The maned wolf Chrysocyon brachyurus Illiger, 1811 is the largest South American canid with a broad distribution; however, the largest portion of its range is in the Brazilian Cerrado savannah, where due to intensive agricultural expansion, it is threatened by habitat loss. Maned wolf population trends are virtually unknown. We analyzed radio telemetry data from a 13-year study in Emas National Park, central Brazil, with Burnham`s live recapture/dead recovery models in the program MARK to obtain the first analytically sound estimate of the apparent maned wolf survival rate. We constructed 16 candidate models including variation in survival rate and resighting probability associated with an individual`s sex or age and year of study. Apparent adult survival rate throughout the study ranged from 0.28 (se=0.08) to 0.97 (se=0.06). There was no evidence for sex specificity but strong support for time variation. Model weights supported an age effect and the subadult survival rate was 0.63 (se=0.15). Results indicate similar life patterns for male and female maned wolves and similar mortality risks for adults and subadults in the study area. The observed temporal fluctuations of adult survival rate are important for population dynamics as they decrease average population growth rates. Population dynamics are central for conservation planning and our results are an important step towards a better understanding of the maned wolf`s ecology.

Effect of remifentanil hydrochloride administered via constant rate infusion on the minimum alveolar concentration of isoflurane in cats

Objective-To evaluate the effects of increasing doses of remifentanil hydrochloride administered via constant rate infusion (CRI) on the minimum alveolar concentration (MAC) of isoflurane in cats. Animals-6 healthy adult cats. Procedures-For each cat, 2 experiments were performed (2-week interval). On each study day, anesthesia was induced and maintained with isoflurane; a catheter was placed in a cephalic vein for the administration of lactated Ringer`s solution or remifentanil CRIs, and a catheter was placed in the jugular vein for collection of blood samples for blood gas analyses. On the first study day, individual basal MAC (MAC(Basal)) was determined for each cat. On the second study day, 3 remifentanil CRIs (0.25, 0.5, and 1.0 mu g/kg/min) were administered (in ascending order); for each infusion, at least 30 minutes elapsed before determination of MAC (designated as MAC(R0.25`) MAC(R0.5`) and MACR(R1.0`) respectively). A 15-minute washout period was allowed between CRIs. A control MAC (MAC Control) was determined after the last remifentanil infusion. Results-Mean +/- SD MAC(Basal) and MAC(Control) values at sea level did not differ significantly (1.66 +/- 0.08% and 1.52 +/- 0.21%, respectively). The MAC values determined for each remifentanil CRI did not differ significantly. However, MACR(0.25`) MAC(R0.5`) and MAC(R1.0) were significantly decreased, compared with MAC(Basal`) by 23.4 +/- 79%, 29.8 +/- 8.3%, and 26.0 +/- 9.4%, respectively. Conclusions and Clinical Relevance-The 3 doses of remifentanil administered via CRI resulted in a similar degree of isoflurane MAC reduction in adult cats, indicating that a ceiling effect was achieved following administration of the lowest dose. (Am J Vet Res 2009;70:581-588)

Differences in thermoregulatory ability between slick-haired and wild-type lactating Holstein cows in response to acute heat stress

Animals inheriting the slick hair gene have a short, sleek, and sometimes glossy coat. The objective of the present study was to determine whether slick-haired Holstein cows regulate body temperature more effectively than wild-type Holstein cows when exposed to an acute increase in heat stress. Lactating slick cows (n = 10) and wild-type cows (n = 10) were placed for 10 h in an indoor environment with a solid roof, fans, and evaporative cooling or in an outdoor environment with shade cloth and no fans or evaporative cooling. Cows were exposed to both environments in a single reversal design. Vaginal temperature, respiration rate, surface temperature, and sweating rate were measured at 1200, 1500, 1800, and 2100 h (replicate 1) or 1200 and 1500 h (replicate 2), and blood samples were collected for plasma cortisol concentration. Cows in the outdoor environment had higher vaginal and surface temperatures, respiration rates, and sweating rates than cows in the indoor environment. In both environments, slick-haired cows had lower vaginal temperatures (indoor: 39.0 vs. 39.4 degrees C; outdoor 39.6 vs. 40.2 degrees C; SEM = 0.07) and respiration rate (indoor: 67 vs. 79 breaths/min; outdoor 97 vs. 107 breaths/min; SEM = 5.5) than wild-type cows and greater sweating rates in unclipped areas of skin (indoor: 57 vs. 43 g.h(-1)/m(2); outdoor 82 vs. 61 g.h(-1)/m(2); SEM = 8). Clipping the hair at the site of sweating measurement eliminated the difference between slick-haired and wild-type cows. Results indicate that slick-haired Holstein cows can regulate body temperature more effectively than wild-type cows during heat stress. One reason slick-haired animals are better able to regulate body temperature is increased sweating rate.