Euler(p,q) Processes and Their Application to Non Stationary Time Series with Time Varying Frequencies

We introduce Euler(p, q) processes as an extension of the Euler(p) processes for purposes of obtaining more parsimonious models for non stationary processes whose periodic behavior changes approximately linearly in time. The discrete Euler(p, q) models are a class of multiplicative stationary (M-stationary) processes and basic properties are derived. The relationship between continuous and discrete mixed Euler processes is shown. Fundamental to the theory and application of Euler(p, q) processes is a dual relationship between discrete Euler(p, q) processes and ARMA processes, which is established. The usefulness of Euler(p, q) processes is examined by comparing spectral estimation with that obtained by existing methods using both simulated and real data.     

Those Who are Left Behind: An Estimate of the Number of Family Members of Suicide Victims in Japan

This paper contributes to the literature of suicide studies by presenting procedures and its estimates of the number of family members who lose their loved ones to suicide. Using Japanese aggregate level data, three main findings emerge: first, there are approximately five bereaved family members per suicide; second, in 2006, there were about 90,000 children who had lost a parent to suicide; and third, in 2006, there were about three million living family members who had lost a loved one to suicide. The direct production loss of bereaved family members in 2006 alone is estimated at approximately 197 million USD. These results are valuable in evaluating the cost-effectiveness of suicide prevention programs and in designing appropriate policy instruments.     

Simulating the average run length for cusum schemes using variance reduction techniques

Some variance reduction techniques utilizing the total hazard are developed to estimate the average run lengths of the cumulative sum charts through simulation when the process follows a general probability distribution. Particularly, we propose the hazard estimator and the cycle estimator. Simulation results are shown for the exponential case and these estimators are compared with the raw simulation estimator. Applicability to multivariate CUSUM schemes is briefly discussed in the conclusion.     

DISCONTINUITY OF OUTPUT CONVERGENCE WITHIN THE UNITED STATES: WHY HAS THE COURSE CHANGED?

Has the progress of output convergence changed within the United States? This article examines the output convergence among U.S. states for the last five decades by making several improvements over the extant literature. By applying a battery of convergence tests designed to capture nonlinear transitional dynamics to real output per worker data (i.e., nominal values deflated by state‐level price), we find that output convergence has not been a feature of the continental United States since the 1970s. Instead, output convergence has proceeded among four subgroups within which constituent states have certain characteristics in common. Our regression analysis suggests that state‐level characteristics related to technology and human capital play a crucial role in accounting for the formation and composition of convergence clubs, in agreement with the recent theoretical models of growth and development (e.g., Aghion et al. 2009; Gennaioli et al. 2013b). The level of technology, proxied by patents, turns out to be a consistently significant determinant even after controlling for endogeneity, suggesting that frictions in the diffusion of technology and human capital may have led to clustering of states with different levels of productivity. Our results therefore cast doubt on the common view that diffusion of knowledge and technology across state borders is frictionless. (JEL O47, O51)     

A Partially Linear Model Using a Gaussian Process Prior

A partially linear model is a semiparametric regression model that consists of parametric and nonparametric regression components in an additive form. In this article, we propose a partially linear model using a Gaussian process regression approach and consider statistical inference of the proposed model. Based on the proposed model, the estimation procedure is described by posterior distributions of the unknown parameters and model comparisons between parametric representation and semi- and nonparametric representation are explored. Empirical analysis of the proposed model is performed with synthetic data and real data applications.     

Social Isolation and Spousal Violence: Comparing Female Marriage Migrants With Local Women

This research examined the impact of network participation, social support, and social control on the violence victimization of female marriage migrants by a spouse. Data were from a household survey of 492 cross‐border and 379 local married couples in Hong Kong in 2007. The findings indicated that female marriage migrants were more vulnerable to spousal violence and more socially isolated, compared with local women. Husbands of marriage migrants were also more socially isolated compared with men who married locally. The network participation of both husband and wife was associated positively with spousal violence against women, whereas their perceived social control was associated negatively with violence. Female marriage migrants' greater vulnerability to spousal physical assault was a result of their younger age compared with local women, but their greater vulnerability to spousal psychological aggression and sexual coercion was related, in part, to their own and their husbands' network participation as well as their level of social control.