Bayesian model comparison with un-normalised likelihoods

Models for which the likelihood function can be evaluated only up to a parameter-dependent unknown normalizing constant, such as Markov random field models, are used widely in computer science, statistical physics, spatial statistics, and network analysis. However, Bayesian analysis of these models using standard Monte Carlo methods is not possible due to the intractability of their likelihood functions. Several methods that permit exact, or close to exact, simulation from the posterior distribution have recently been developed. However, estimating the evidence and Bayes’ factors for these models remains challenging in general. This paper describes new random weight importance sampling and sequential Monte Carlo methods for estimating BFs that use simulation to circumvent the evaluation of the intractable likelihood, and compares them to existing methods. In some cases we observe an advantage in the use of biased weight estimates. An initial investigation into the theoretical and empirical properties of this class of methods is presented. Some support for the use of biased estimates is presented, but we advocate caution in the use of such estimates.

Sampling uncertainty in gridded sea surface temperature products and Advanced Very High Resolution Radiometer (AVHRR) Global Area Coverage (GAC) data

Sea surface temperature (SST) data are often provided as gridded products, typically at resolutions of order 0.05 degrees from satellite observations to reduce data volume at the request of data users and facilitate comparison against other products or models. Sampling uncertainty is introduced in gridded products where the full surface area of the ocean within a grid cell cannot be fully observed because of cloud cover. In this paper we parameterise uncertainties in SST as a function of the percentage of clear-sky pixels available and the SST variability in that subsample. This parameterisation is developed from Advanced Along Track Scanning Radiometer (AATSR) data, but is applicable to all gridded L3U SST products at resolutions of 0.05-0.1 degrees, irrespective of instrument and retrieval algorithm, provided that instrument noise propagated into the SST is accounted for. We also calculate the sampling uncertainty of ~0.04 K in Global Area Coverage (GAC) Advanced Very High Resolution Radiometer (AVHRR) products, using related methods.

English compound and non-compound processing in bilingual and multilingual speakers: effects of dominance and sequential multilingualism

This article reports on a study investigating the relative influence of the first and dominant language on L2 and L3 morpho-lexical processing. A lexical decision task compared the responses to English NV-er compounds (e.g., taxi driver) and non-compounds provided by a group of native speakers and three groups of learners at various levels of English proficiency: L1 Spanish-L2 English sequential bilinguals and two groups of early Spanish-Basque bilinguals with English as their L3. Crucially, the two trilingual groups differed in their first and dominant language (i.e., L1 Spanish-L2 Basque vs. L1 Basque-L2 Spanish). Our materials exploit an (a)symmetry between these languages: while Basque and English pattern together in the basic structure of (productive) NV-er compounds, Spanish presents a construction that differs in directionality as well as inflection of the verbal element (V[3SG] + N). Results show between and within group differences in accuracy and response times that may be ascribable to two factors besides proficiency: the number of languages spoken by a given participant and their dominant language. An examination of response bias reveals an influence of the participants' first and dominant language on the processing of NV-er compounds. Our data suggest that morphological information in the nonnative lexicon may extend beyond morphemic structure and that, similarly to bilingualism, there are costs to sequential multilingualism in lexical retrieval.

Impact of lipoprotein lipase gene polymorphism, S447X, on postprandial triacylglycerol and glucose response to sequential meal ingestion

Lipoprotein lipase (LPL) is a key rate-limiting enzyme for the hydrolysis of triacylglycerol (TAG) in chylomicrons and very low-density lipoprotein. Given that postprandial assessment of lipoprotein metabolism may provide a more physiological perspective of disturbances in lipoprotein homeostasis compared to assessment in the fasting state, we have investigated the influence of two commonly studied LPL polymorphisms (rs320, HindIII; rs328, S447X) on postprandial lipaemia, in 261 participants using a standard sequential meal challenge. S447 homozygotes had lower fasting HDL-C (p = 0.015) and a trend for higher fasting TAG (p = 0.057) concentrations relative to the 447X allele carriers. In the postprandial state, there was an association of the S447X polymorphism with postprandial TAG and glucose, where S447 homozygotes had 12% higher TAG area under the curve (AUC) (p = 0.037), 8.4% higher glucose-AUC (p = 0.006) and 22% higher glucose-incremental area under the curve (IAUC) (p = 0.042). A significant gene–gender interaction was observed for fasting TAG (p = 0.004), TAG-AUC (Pinteraction = 0.004) and TAG-IAUC (Pinteraction = 0.016), where associations were only evident in men. In conclusion, our study provides novel findings of an effect of LPL S447X polymorphism on the postprandial glucose and gender-specific impact of the polymorphism on fasting and postprandial TAG concentrations in response to sequential meal challenge in healthy participants