ABSTRACT
We propose a unified framework for likelihood-based regression modeling when the response variable has finite support. Our work is motivated by the fact that, in practice, observed data are discrete and bounded. The proposed methods assume a model which includes models previously considered for interval-censored variables with log-concave distributions as special cases. The resulting log-likelihood is concave, which we use to establish asymptotic normality of its maximizer as the number of observations n tends to infinity with the number of parameters d fixed, and rates of convergence of L1 -regularized estimators when the true parameter vector is sparse and d and n both tend to infinity with log ( d ) / n â 0 $\log (d) / n \rightarrow 0$ . We consider an inexact proximal Newton algorithm for computing estimates and give theoretical guarantees for its convergence. The range of possible applications is wide, including but not limited to survival analysis in discrete time, the modeling of outcomes on scored surveys and questionnaires, and, more generally, interval-censored regression. The applicability and usefulness of the proposed methods are illustrated in simulations and data examples.
Subject(s)
Algorithms , Likelihood Functions , Regression Analysis , Computer Simulation , Survival AnalysisABSTRACT
Femtosecond transient absorption pump-probe spectroscopy has been used to study the third- and fifth-order nonlinear susceptibility of UV-grade fused silica. Theoretical simulations support the experimental evidence of interference between the third- and fifth-order contributions to the signal. Approximate values of the second-order nonlinear refractive index n(4) and of the three-photon absorption coefficient gamma are obtained.
ABSTRACT
I investigated the occurrence, frequency and distribution of allosuckling in a wild population of fallow deer, Dama dama, throughout the lactation period during 1992 in southern Sweden. A total of 292 suckling bouts were observed in four groups; in 43% of these, fawns were seen sucking from a female that was not the mother. Allosuckling occurred in a higher percentage of suckling bouts as the lactation period progressed and all 16 fawns participated to various degrees. Of the 16 females, 13 suckled nonoffspring fawns but their behaviour towards these fawns varied greatly. Allosuckling was positively correlated with the age of the females and negatively correlated with the rate of their aggressive behaviour. Since allosuckling was not performed between groups but was common between related members in a stable social unit and was expressed reciprocally between the oldest individuals in the group, allosuckling could be a result of kin selection and/or reciprocal altruism. Copyright 1998 The Association for the Study of Animal Behaviour.
ABSTRACT
To find the causes of faster growth in males, we studied the growth and behaviour of six male and five female fallow deer, Dama dama, fawns, hand-reared on the same amount and quality of milk, and compared them with naturally reared fawns. The bottle-reared fawns grew more slowly. In particular, the difference in weight gain between bottle- and mother-reared fawns was significantly larger for males during the time when they consumed only milk. Male fawns also sucked harder and were more motivated to obtain milk. These results indicate that male fallow deer fawns receive more milk from their mothers than female fawns under natural conditions and, hence, maternal investment seems male biased in this species. Males grew slightly, but significantly, faster until 10 weeks old, which revealed that sex-differential growth during the period of maternal investment is possible in the absence of differential maternal investment. The results could not reveal whether this small difference in growth is brought about by physiological or behavioural differences between the sexes.Copyright 1998 The Association for the Study of Animal Behaviour
