Hierarchical likelihood approach to non-Gaussian factor analysis

Factor models, structural equation models (SEMs) and random-effect models share the common feature that they assume latent or unobserved random variables. Factor models and SEMs allow well developed procedures for a rich class of covariance models with many parameters, while random-effect models allow well developed procedures for non-normal models including heavy-tailed distributions for responses and random effects. In this paper, we show how these two developments can be combined to result in an extremely rich class of models, which can be beneficial to both areas. A new fitting procedures for binary factor models and a robust estimation approach for continuous factor models are proposed.     

Revisiting the effectiveness of the employment‐oriented welfare state: Considering the quality of employment achievement

The purpose of this study was to examine the effectiveness of an employment‐oriented welfare state. While previous studies have focused on employment growth, this study considers the quality of employment, especially its mediating impact between public social expenditure and fiscal soundness. Three‐step mediated regression analysis was conducted on a data sample from 19 developed countries from 1991 to 2013. The results show that while public social expenditure negatively affects fiscal soundness, if it promotes total and part‐time employment, it can positively affect fiscal soundness. However, when the incidence of involuntary part‐time employment is high, public social expenditure and its impact on employment performance no longer guarantee fiscal soundness. This study addresses the importance of promoting not only employment growth but also employment quality to secure fiscal soundness.     

Quantile Regression for Location‐Scale Time Series Models with Conditional Heteroscedasticity

This paper considers quantile regression for a wide class of time series models including autoregressive and moving average (ARMA) models with asymmetric generalized autoregressive conditional heteroscedasticity errors. The classical mean‐variance models are reinterpreted as conditional location‐scale models so that the quantile regression method can be naturally geared into the considered models. The consistency and asymptotic normality of the quantile regression estimator is established in location‐scale time series models under mild conditions. In the application of this result to ARMA‐generalized autoregressive conditional heteroscedasticity models, more primitive conditions are deduced to obtain the asymptotic properties. For illustration, a simulation study and a real data analysis are provided.     

Dealing with the Initial Observation in the LM Unit Root Test

This article proposes a linearly weighted unit root test with a new weighting scheme which reflects the trade-off in power between the ADF and LM tests regarding the initial value of a time series. Simulation results indicate that the proposed test has better power performance and works better than other available tests in the literature for a range of initial conditions.     

Mad, Bad, or Reasonable? Newspaper Portrayals of the Battered Woman Who Kills

A heated debate about battered women who kill abusive male partners started in the 1970s. In this study, we tracked the public discourse on battered women who kill by coding 250 newspaper articles published between 1978 and 2002. Using four typifying models, we found that leading explanations for why battered women kill medicalized then criminalized their actions; they were mad then bad. We also found that reporters used quotes from claims makers supporting conventional or medical typifications of battered women to a much greater degree than statements from alternative, feminist sources. In conclusion, simplified, sensational and conventional understandings of crime causation drove the social construction of “the battered woman who kills”. She may be mad or bad, but rarely has she been portrayed as reasonable. Suggestions for promoting feminist narrative in the media are also provided.     

H-likelihood: problems and solutions

In recent issues of this journal it has been asserted in two papers that the use of h-likelihood is wrong, in the sense of giving unsatisfactory estimates of some parameters for binary data (Kuk and Cheng, 1999; Waddington and Thompson, 2004) or theoretically unsound (Kuk and Cheng, 1999). We wish to refute both these assertions.     

Do we need the constant term in the heterogenous autoregressive model for forecasting realized volatilities?

No-constant strategy is considered for the heterogenous autoregressive (HAR) model of Corsi, which is motivated by smaller biases of its estimated HAR coefficients than those of the constant HAR model. The no-constant model produces better forecasts than the constant model for four real datasets of the realized volatilities (RVs) of some major assets. Robustness of forecast improvement is verified for other functions of realized variance and log RV and for the extended datasets of all 20 RVs of Oxford-Man realized library. A Monte Carlo simulation also reveals improved forecasts for some historic HAR model estimated by Corsi.     

Does unemployment increase suicide rates? The OECD panel evidence

Previous studies of whether unemployment increases suicide rates give mixed results. None of them controlled for an interaction between unemployment and income. This paper tests the hypothesis whether the relationship between unemployment rates and suicide rates vary according to the level of real per capita GDP. We use the cross-country panel fixed effects approach to exclude cross-sectional variations but exploit time-series ones. We support that higher income is associated with higher suicide rates. In particular, the evidence shows that the implied effect of unemployment on suicide rates is positive for countries with higher income. Actually, for countries with lower-income levels, there is a negative impact of unemployment on suicides.     

An RKHS framework for functional data analysis

Linear combinations of random variables play a crucial role in multivariate analysis. Two extension of this concept are considered for functional data and shown to coincide using the Loève–Parzen reproducing kernel Hilbert space representation of a stochastic process. This theory is then used to provide an extension of the multivariate concept of canonical correlation. A solution to the regression problem of best linear unbiased prediction is obtained from this abstract canonical correlation formulation. The classical identities of Lawley and Rao that lead to canonical factor analysis are also generalized to the functional data setting. Finally, the relationship between Fisher's linear discriminant analysis and canonical correlation analysis for random vectors is extended to include situations with function-valued random elements. This allows for classification using the canonical Y scores and related distance measures.