Split Hamiltonian Monte Carlo

We show how the Hamiltonian Monte Carlo algorithm can sometimes be speeded up by “splitting” the Hamiltonian in a way that allows much of the movement around the state space to be done at low computational cost. One context where this is possible is when the log density of the distribution of interest (the potential energy function) can be written as the log of a Gaussian density, which is a quadratic function, plus a slowly-varying function. Hamiltonian dynamics for quadratic energy functions can be analytically solved. With the splitting technique, only the slowly-varying part of the energy needs to be handled numerically, and this can be done with a larger stepsize (and hence fewer steps) than would be necessary with a direct simulation of the dynamics. Another context where splitting helps is when the most important terms of the potential energy function and its gradient can be evaluated quickly, with only a slowly-varying part requiring costly computations. With splitting, the quick portion can be handled with a small stepsize, while the costly portion uses a larger stepsize. We show that both of these splitting approaches can reduce the computational cost of sampling from the posterior distribution for a logistic regression model, using either a Gaussian approximation centered on the posterior mode, or a Hamiltonian split into a term that depends on only a small number of critical cases, and another term that involves the larger number of cases whose influence on the posterior distribution is small.     

Caloric restriction minimizes aging effects on the femoral medial condyle

This study aimed to analyze the effects of caloric restriction on aged femoral articular cartilage of Wistar rats. Three groups of eight animals each were considered: young (YC) and old (OC) control groups fed with a normal diet and old caloric restriction group (OCR) composed of 18-month-old animals fed with a 31% less caloric diet from 6-months of age. Articular cartilage was studied through morphometry and immunohistochemistry. Body mass was 12% less in the OCR group than in the OC group. The articular cartilage from OC rats show thinner medial condyles, fewer chondrocytes, smaller chondrocytes nuclear volume and, in both condyles, a predominance of collagen type II and less collagen density compared to both YC and OCR groups (p?p?相似文献   

The Detection of Influential Observations for Allocation,Separation, and the Determination of Probabilities in a Bayesian Framework

Normal theory separation and allocation problems are discussed from a predictive point of view. Influence statistics are defined and employed to ascertain the impact that particular observations will have on the inferential goals—allocation of future observations, separation between populations, and the determination of probabilities for future cases. Methods are illustrated on a collection of financial data taken from Johnson and Wichern (1982).     

Dual experiences of teenage parenting and homelessness among Native Hawaiian youth: A critical interpretative phenomenological analysis

Teenage parenting is a challenging venture—particularly when parenting while homeless. The present study explores in‐depth through case‐based analysis the lived experiences of three young homeless and parenting teens (one mother and two fathers). It is the first known study of teenage parenting and homelessness among Native Hawaiian youth, a key affected population. Interpretative phenomenological analysis of participants' individual interviews revealed three themes that characterize the experience of parenting as a teen while homeless: (a) They never raised me up (childhood experiences); (b) becoming mommy and daddy (early experiences of teen pregnancy and parenting while homeless); and (c) finding our place (current experiences of teen parenting and perspectives on teen pregnancy and homelessness). Implications for direct practice, social policy, and future research are discussed in the context of this complex reality.     

The time and the transitions back to work in France after maternity

Review of Economics of the Household - Previous studies have shown that childrearing has a different impact on a mother’s professional career, depending, among other reasons, on how much time...     

Does the capital of social capital matter? Relational resources of the board and the performance of Brazilian companies

Based on theories of social capital, in this study, we seek to assess the impact of a board’s social capital on the market value of companies listed on the Brazilian stock exchange. As our indicator of social capital, we use the relational resources identified in the direct, indirect and heterogeneous ties of the board. Employing panel data from 508 observations, our results indicate that heterogeneous relational resources have a stronger and more significant influence than the resources available from board members’ direct relationships. Additionally, as the effects of board interlock are endogenously determined by several factors related to the firm level, we seek to mitigate the endogeneity problem using models of instrumental variables and simultaneous equations. Our hypotheses were consistent after controlling for endogeneity. We also check whether the board’s social capital could present a U-inverted effect on the market value. This relationship was only plausible in social capital by indirect ties. Finally, we isolate the effect of relational resources within and between industries on Tobin’s Q. There was no significant effect through interlocks within the same industry. However, ties with companies in several other industries were significant.