Sparse Logistic Regression: Comparison of Regularization and Bayesian Implementations

In knowledge-based systems, besides obtaining good output prediction accuracy, it is crucial to understand the subset of input variables that have most influence on the output, with the goal of gaining deeper insight into the underlying process. These requirements call for logistic model estimation techniques that provide a sparse solution, i.e., where coefficients associated with non-important variables are set to zero. In this work we compare the performance of two methods: the first one is based on the well known Least Absolute Shrinkage and Selection Operator (LASSO) which involves regularization with an ℓ 1 norm; the second one is the Relevance Vector Machine (RVM) which is based on a Bayesian implementation of the linear logistic model. The two methods are extensively compared in this paper, on real and simulated datasets. Results show that, in general, the two approaches are comparable in terms of prediction performance. RVM outperforms the LASSO both in term of structure recovery (estimation of the correct non-zero model coefficients) and prediction accuracy when the dimensionality of the data tends to increase. However, LASSO shows comparable performance to RVM when the dimensionality of the data is much higher than number of samples that is p > > n .

Speech Perception: Phonological Neighborhood Effects on Word Recognition Persist Despite Semantic Sentence Context

This study tested the hypothesis that two lexical properties, both phonological neighborhood density (ND) and neighborhood frequency (NF), influence the recognition of target words when preceded by either a semantically congruent or semantically neutral context. Our study is the first to test this hypothesis using a language other than English (i.e., Spanish). We used highly familiar bisyllabic nouns with medium-frequency occurrence as target words, and we expected recognition accuracy to increase as ND and NF decreased in both semanticallly congruent and semantically neutral sentences. We presented 48 undergraduate listeners with a set of 80 words, differing in ND and NF, within these two sentence contexts (i.e., 160 sentences). We then tested the relationships between ND, NF, and variations in semantic sentence context within a linear logistic model and found that words with a low frequency of neighbors were more likely to be correctly recognized in both sentence contexts. Thus, during word recognition, the influence of phonological competition outweighed semantic sentence context even when words were presented in Spanish.

Using JAGS for Bayesian Cognitive Diagnosis Modeling: A Tutorial

In this article, we systematically introduce the just another Gibbs sampler (JAGS) software program to fit common Bayesian cognitive diagnosis models (CDMs) including the deterministic inputs, noisy “and” gate model; the deterministic inputs, noisy “or” gate model; the linear logistic model; the reduced reparameterized unified model; and the log-linear CDM (LCDM). Further, we introduce the unstructured latent structural model and the higher order latent structural model. We also show how to extend these models to consider polytomous attributes, the testlet effect, and longitudinal diagnosis. Finally, we present an empirical example as a tutorial to illustrate how to use JAGS codes in R.

An inverse-linear logistic model of the main sequence

A model of the main sequence is proposed based on the logistic function. The model’s fit to the peak velocity-amplitude relation resembles an S curve, simulta- neously allowing control of the curve’s asymptotes at very small and very large amplitudes, as well as its slope over the mid amplitude range. The proposed inverse-linear logistic model is also able to express the linear relation of duration and amplitude. We demonstrate the utility and robustness of the model when fit to aggregate data at the small- and mid-amplitude ranges, namely when fitting microsaccades, saccades, and superposition of both. We are confident the model will suitably extend to the large-amplitude range of eye movements.

Non-linear disease tolerance curves reveal distinct components of host responses to viral infection

The ability to tolerate infection is a key component of host defence and offers potential novel therapeutic approaches for infectious diseases. To yield successful targets for therapeutic intervention, it is important that the analytical tools employed to measure disease tolerance are able to capture distinct host responses to infection. Here, we show that commonly used methods that estimate tolerance as a linear relationship may be inadequate, and that more flexible, non-linear estimates of this relationship may reveal variation in distinct components of host defence. To illustrate this, we measured the survival of Drosophila melanogaster carrying either a functional or non-functional regulator of the JAK-STAT immune pathway (G9a) when challenged with a range of concentrations of Drosophila C Virus (DCV). While classical linear model analyses indicated that G9a affected tolerance only in females, a more powerful non-linear logistic model showed that G9a mediates viral tolerance to different extents in both sexes. This analysis also revealed that G9a acts by changing the sensitivity to increasing pathogen burdens, but does not reduce the ultimate severity of infection. These results indicate that fitting non-linear models to host health-pathogen burden relationships may offer better and more detailed estimates of disease tolerance.

Plasma Insulin-Like Growth Factor 1 Is Associated with Cognitive Impairment in Parkinson's Disease

Objective: The aim of this study was to test plasma insulin-like growth factor 1 (IGF-1) change in Parkinson's disease (PD) and essential tremor (ET), and assess the association of plasma IGF-1 level with motor and nonmotor symptoms in PD and ET. Methods: Plasma IGF-1 was measured in 100 PD patients, 40 ET patients, and 76 healthy controls. Motor and nonmotor symptoms were assessed by different scales. Spearman correlation test and linear logistic model were used to analyze the correlation of plasma IGF-1 with motor and nonmotor symptoms of PD and ET. Results: The plasma IGF-1 level was significantly increased in PD compared to healthy controls and ET patients. In addition, low plasma IGF-1 was correlated with low Mini-Mental State Examination (MMSE) scores in PD patients. However, no correlation was found between plasma IGF-1 and MMSE scores in ET patients. Conclusion: Plasma IGF-1 increased significantly in PD but remained unchanged in ET. A low plasma IGF-1 level was associated with poor cognitive performance in PD but not in ET patients.

SEX RATIO AT BIRTH AND FAMILY COMPOSITION IN SUB-SAHARAN AFRICA: INTER-COUPLE VARIATIONS

SummaryIn this study, sex ratios at birth (male/female births) were found to vary according to family composition. Using Demographic and Health Survey (DHS) maternity histories from sub-Saharan Africa, the study shows that the sex ratio at birth increases with the number of previous male births and decreases with the number of previous female births. For families with only males, the sex ratio increases from 1·046 for the first birth to 1·083 for the 8th birth. For families with only females, the sex ratio decreases from 1·046 for the first birth to 0·942 for the 8th birth. The differences were highly significant when tested with a linear logistic model (p=0·018 for males; p=1·85✕10−11 for females). The effect was not symmetrical, and was found to be significantly stronger for females. These effects could be reproduced assuming a strong heterogeneity between couples. The distribution of sex ratios was fitted with an asymmetrical log-gamma function, which revealed a wide range of variation in the sex ratio between 0·50 and 1·30, and a peak around 1·14. The results and their implications are discussed in the light of former findings in France and in the United States of America.