Study of partial least squares and ridge regression methods

This article considers both Partial Least Squares (PLS) and Ridge Regression (RR) methods to combat multicollinearity problem. A simulation study has been conducted to compare their performances with respect to Ordinary Least Squares (OLS). With varying degrees of multicollinearity, it is found that both, PLS and RR, estimators produce significant reductions in the Mean Square Error (MSE) and Prediction Mean Square Error (PMSE) over OLS. However, from the simulation study it is evident that the RR performs better when the error variance is large and the PLS estimator achieves its best results when the model includes more variables. However, the advantage of the ridge regression method over PLS is that it can provide the 95% confidence interval for the regression coefficients while PLS cannot.     

A note on the non-negativity of continuous-time ARMA and GARCH processes

A general approach for modeling the volatility process in continuous-time is based on the convolution of a kernel with a non-decreasing Lévy process, which is non-negative if the kernel is non-negative. Within the framework of Continuous-time Auto-Regressive Moving-Average (CARMA) processes, we derive a necessary condition for the kernel to be non-negative, and propose a numerical method for checking the non-negativity of a kernel function. These results can be lifted to solving a similar problem with another approach to modeling volatility via the COntinuous-time Generalized Auto-Regressive Conditional Heteroscedastic (COGARCH) processes.     

Robust williams trent test

To investigate the biological activities of a new compound or drug, experimenters usually compare a series of increasing doses to a control. Among other objectives, one may try to investigate any possible dose-response trend and to determine the minimum effective dose among all the experimental doses. Williams (1971, 1972) proposed a procedure to test the dose-response trend and also to identify the minimum effective dose based on the normally distributed data. In this paper, we propose a similar test procedure based on the robust estimate'of the average response to perform similar analysis. The proposed method is more resistant to the outliers and more powerful than the Williams procedure when the data distribution deviates from normality. We illustrate the use of this procedure with data arising from a recent study.     

A correlation type procedure for assessing multivariate normality

A correlation-type statistic for assessing multivariate normality is described. Its estimated finite sample distribution is tabulated, and its performance against certain alternatives is compared with that of a competing Cramer-von Mises type statistic in a Monte Carlo power study. A set of quadrivariate data is examined as illustration of the procedure.     

Psychometric testing of the Chinese version of ISPCAN Child Abuse Screening Tools Children's Home Version (ICAST-CH-C)

BackgroundChild maltreatment is a global problem and the true extent remains unknown. The International Society for the Prevention of Child Abuse and Neglect (ISPCAN) Child Abuse Screening Tool — Children's Home version (ICAST-CH) has provided accurate assessment of the scope and prevalence of child maltreatment. Yet measures of children's experiences of child maltreatment are limited in the Chinese population.ObjectivesThe study aimed to translate and validate a Chinese version of the ISPCAN Child Abuse Screening Tool — Children's Home version (ICAST-CH) and to evaluate its reliability and validity among Taiwan adolescents.MethodsA three phase study was conducted. In phase 1, the ICAST was translated into Chinese using forward–backward translation procedures with the translation equivalence and content validity assessed. In phase 2, the data provided by a convenience sample of 98 adolescents was used to assess the internal consistency of the ICAST-CH Chinese version (ICAST-CH-C). In phase 3, the psychometric properties of the ICAST-CH-C were tested with a nationwide random sample of 5236 adolescents from 35 schools.ResultsThe translation equivalence and content validity index of the ICAST-CH-C was satisfactory. The inter-rater agreements were .90–.91 for comparability of language and .89–.94 for similarity of interpretability. Results indicated that the ICAST-CH-C had a high level of equivalence with the original English version and demonstrated a high internal consistency (.71–.89). Confirmatory factor analysis revealed the presence of five factors supporting the conceptual dimension of the original instrument.ConclusionThis study provided initial psychometric properties of the ICAST-CH-C and supports it as a reliable, valid, and highly usable instrument to identify childhood victimization in adolescents. It provided health care professionals with a useful tool to assess the severity and prevalence of child maltreatment within Chinese communities.     

Smoothing parameter selection in nonparametric regression using an improved Akaike information criterion

Many different methods have been proposed to construct nonparametric estimates of a smooth regression function, including local polynomial, (convolution) kernel and smoothing spline estimators. Each of these estimators uses a smoothing parameter to control the amount of smoothing performed on a given data set. In this paper an improved version of a criterion based on the Akaike information criterion (AIC), termed AIC_C, is derived and examined as a way to choose the smoothing parameter. Unlike plug-in methods, AIC_C can be used to choose smoothing parameters for any linear smoother, including local quadratic and smoothing spline estimators. The use of AIC_C avoids the large variability and tendency to undersmooth (compared with the actual minimizer of average squared error) seen when other 'classical' approaches (such as generalized cross-validation (GCV) or the AIC) are used to choose the smoothing parameter. Monte Carlo simulations demonstrate that the AIC_C-based smoothing parameter is competitive with a plug-in method (assuming that one exists) when the plug-in method works well but also performs well when the plug-in approach fails or is unavailable.     

ON THE INTERNAL REVENUE SERVICE'S SERVICE AND ENFORCEMENT

Built on the seminal work of Graetz et al. (1986), this study considers a model of tax compliance in which the Internal Revenue Service (IRS) services as well as audits taxpayers as in the real world. Taxpayers are uncertain about whether their incomes are eligible for deduction/exemption, and may seek advice from the taxpayer service provided by the IRS to mitigate the uncertainty. We show: (1) while the society's preferred quality of taxpayer service is likely to be sufficiently high, the IRS's preferred quality of taxpayer service is only at an intermediate level; and (2) the conflict between the society and the IRS tends to be intensified as the cost of tax audits becomes lower. Evidence in support of our results is addressed. We also extend our model to addressing the issue of corruption in tax administration and to considering a dynamic model. (JEL D82, H26, K42)     

Hedging Barrier Options in GARCH Models with Transaction Costs

This study proposes a modified strike‐spread method for hedging barrier options in generalized autoregressive conditional heteroskedasticity (GARCH) models with transaction costs. A simulation study was conducted to investigate the hedging performance of the proposed method in comparison with several well‐known static methods for hedging barrier options. An accurate, easy‐to‐implement and fast scheme for generating the first passage time under the GARCH framework which enhances the accuracy and efficiency of the simulation is also proposed. Simulation results and an empirical study using real data indicate that the proposed approach has a promising performance for hedging barrier options in GARCH models when transaction costs are taken into consideration.     

Three provocations for civic crowdfunding

The rapid rise of crowdfunding in the past five years, most prominently among US-based platforms such as Kickstarter and IndieGoGo, has begun to attract the attention of a wide range of scholars, policymakers and practitioners. This paper considers civic crowdfunding – the use of crowdfunding for projects that produce community or quasipublic assets – and argues that its emergence demands a fresh set of questions and approaches. The work draws on critical case studies constructed through fieldwork in the United States, the UK and Brazil, and a discourse analysis of civic crowdfunding projects collected from platforms by the author. It offers three provocations to scholars and practitioners considering the practice, questioning the extent to which civic crowdfunding is participatory, the extent to which it addresses or contributes to social inequality, and the extent to which it augments or weakens the role of public institutions. In doing so, it finds that civic crowdfunding is capable of vastly divergent outcomes, and argues that the extent to which civic crowdfunding produces outcomes that are beneficial, rather than harmful to the public sphere, will be determined by the extent to which the full range of stakeholders in civic life participate in the practice.