Small‐sample testing inference in symmetric and log‐symmetric linear regression models

This paper deals with the issue of testing hypotheses in symmetric and log‐symmetric linear regression models in small and moderate‐sized samples. We focus on four tests, namely, the Wald, likelihood ratio, score, and gradient tests. These tests rely on asymptotic results and are unreliable when the sample size is not large enough to guarantee a good agreement between the exact distribution of the test statistic and the corresponding chi‐squared asymptotic distribution. Bartlett and Bartlett‐type corrections typically attenuate the size distortion of the tests. These corrections are available in the literature for the likelihood ratio and score tests in symmetric linear regression models. Here, we derive a Bartlett‐type correction for the gradient test. We show that the corrections are also valid for the log‐symmetric linear regression models. We numerically compare the various tests and bootstrapped tests, through simulations. Our results suggest that the corrected and bootstrapped tests exhibit type I probability error closer to the chosen nominal level with virtually no power loss. The analytically corrected tests as well as the bootstrapped tests, including the Bartlett‐corrected gradient test derived in this paper, perform with the advantage of not requiring computationally intensive calculations. We present a real data application to illustrate the usefulness of the modified tests.     

Empirical evaluation of educational interactive systems

Educational software systems are usually highly interactive systems that needs to be evaluated from two different perspectives, first as interactive software, mainly using usability and accessibility approaches, and second as an educational product that needs to be evaluated from its learning impact. The empirical evaluation methods may be quantitative, with a more deductive reasoning, or qualitative, with a more inductive reasoning, oriented, however mixed approaches have been used extensively because they combine the both main methods with a more pragmatic knowledge focusing on knowing the consequences or solving problems. The empirical evaluation of interactive educational software due to its own characteristics requires a mixed approach typically. This special issue is devoted to identify, share and valorise best practices and experiences, with a very solid modelling background, that are focused on the empirical evaluation of educational interactive systems with a special emphasis on mixed quantitative and qualitative approaches.     

International transatlantic trade liberalization: zooming in on regional impacts

The current study analyses the impact on a Portuguese small island regional economy of the Transatlantic Trade and Investment Partnership (T-TIP) between the EU and the USA. A dynamic Computational General Equilibrium (CGE) model detailing six household categories, 45 sectors, and four trading partners is used. Previous studies used aggregate variables and, largely, were based on the structure of the national economy. For a small, integrated economy, foreign trade statistics comprise an underestimation, given that most of the trade occurs through national logistics centres. Taking into account the national integration effects, gross domestic value was estimated to be higher than in other studies. Using equivalent variation, the estimated welfare impact is positive for all six household categories. Value-added suffers mixed impacts depending on the sector. It is negative for fisheries, ambiguous for agriculture and positive for tourism and transportation. The contribution of the current study is to highlight the importance of looking beneath the trade block and national conclusions particularly when regional economic policy is relevant as is the case in Europe. Better knowledge of welfare, regional and sector impacts allows for improved development and mitigation policies.     

Happier with the Same: Job Satisfaction of Disadvantaged Workers

Job satisfaction evaluations depend not only on the objective circumstances that workers experience in their jobs, but also on their subjective dispositions, such as their aspirations, expectations, feelings of entitlement or personal evaluation criteria. We use matched employer–employee data from the United Kingdom to examine whether and how subjective dispositions influencing job satisfaction vary across workers with different socio‐demographic traits. We approximate jobs using detailed occupations within workplaces and find that most of the variability in job satisfaction is at the worker rather than the proximate‐job level, and that workers with disadvantaged statuses report higher satisfaction with the same jobs than those with advantaged statuses.     

Seasonality,price discrimination,and welfare

We discuss a potential limitation to a widely accepted result, namely that an output increase is a necessary condition for welfare to increase with price discrimination. We use a theoretical model to show that the existence of seasonal demand fluctuations may allow for a simultaneous reduction in average output and increase in average welfare. We also discuss a number of extensions of our basic model.     

Corporations and local economies in the Brazilian Amazon: the impacts of the mining sectors scheduled investments in Southeastern Pará (2004–2010)

Southeastern Pará is an extraordinarily dynamic region. There were large cattle ranching projects, financed in the 1960s. Expansion of peasant agriculture, mineral projects and gold prospecting followed the ranching projects. Also important structural transformations reinforced the role of the urban centres and the local rural bases in the logistics of new economic sectors conditioned by the presence of Companhia Vale do Rio Doce (CVRD), in the region since 1985. This article presents the results of an input–output analysis of the investment programme of CVRD from year 2004 up to 2010. The main findings are that the mineral sector has achieved a considerable influence over the economy of Southeastern Pará, so that during the cycle of investments each 1% in the growth of the mineral production creates growth possibilities of about 0.83% for the agricultural sector and about 0.86% for the local urban sectors. Considering just the production growth, a percentile point will imply growth at, respectively, 0.73% and 0.76% for those sectors. For the economy of the rest of the state of Pará, those elasticities would be, incorporating the investments, 0.80% and, without them, 0.68%. For the economy of the rest of Brazil they would be, respectively, 0.88% and 0.78%.     

Rivalry within strategic groups and consequences for performance: the firm‐size effects

Our study examines how, in a given industry, rivalry functions within strategic groups defined according to the size of their member firms and how this rivalry affects performance. We hypothesize that, owing to several forms of group‐level effects including market power, efficiency, differentiation, and multimarket contact, strategic groups that comprise smaller firms will exhibit both increased rivalry and decreased performance compared with strategic groups that comprise larger firms. We test our hypotheses by estimating the effect of group‐level strategic interactions (i.e., conjectural variations) on firm performance. Ultimately, our analysis of empirical data on loans in the Spanish banking industry demonstrates that increased rivalry and decreased performance indeed characterizes firms belonging to a strategic group that comprises smaller firms. Copyright © 2011 John Wiley & Sons, Ltd.     

Holt’s exponential smoothing and neural network models for forecasting interval-valued time series

Interval-valued time series are interval-valued data that are collected in a chronological sequence over time. This paper introduces three approaches to forecasting interval-valued time series. The first two approaches are based on multilayer perceptron (MLP) neural networks and Holt’s exponential smoothing methods, respectively. In Holt’s method for interval-valued time series, the smoothing parameters are estimated by using techniques for non-linear optimization problems with bound constraints. The third approach is based on a hybrid methodology that combines the MLP and Holt models. The practicality of the methods is demonstrated through simulation studies and applications using real interval-valued stock market time series.     

Unintended Effects of Targeting an Environmental Rebate

When designing schemes such as conditional cash transfers or payments for ecosystem services, the choice of whom to select and whom to exclude is critical. We incentivize and measure actual contributions to an environmental public good to ascertain whether being excluded from a rebate can affect contributions and, if so, whether the rationale for exclusion influences such effects. Treatments, i.e., three rules that determine who is selected and excluded, are randomly assigned. Two of the rules base exclusion on subjects’ initial contributions. The third is based upon location and the rationales are always explained. The rule that targets the rebate to low initial contributors, who have more potential to raise contributions, is the only rule that raised contributions by those selected. Yet by design, that same rule excludes the subjects who contributed the most initially. They respond by reducing their contributions even though their income and prices are unchanged.