Aspects of statistical analysis in DEA-type frontier models

In Grosskopf (1995) and Banker (1995) different approaches and problems of statistical inference in DEA frontier models are presented. This paper focuses on the basic characteristics of DEA models from a statistical point of view. It arose from comments and discussions on both papers above. The framework of DEA models is deterministic (all the observed points lie on the same side of the frontier), nevertheless a stochastic model can be constructed once a data generating process is defined. So statistical analysis may be performed and sampling properties of DEA estimators can be established. However, practical statistical inference (such as test of hypothesis, confidence intervals) still needs artifacts like the bootstrap to be performed. A consistent bootstrap relies also on a clear definition of the data generating proces and on a consistent estimator of it: The approach of Simar and Wilson (1995) is described. Finally, some trails are proposed for introducing stochastic noise in DEA models, in the spirit of the Kneip-Simar (1995) approach.     

股权分置改革受限股份流通次序

本文采用分量回归模型．具体研究国有股权对不同业绩水平上市公司影响程度的差异。结论表明．国有股权对公司业绩具有负面影响，且对不同业绩水平上市公司的影响程度存在明显差异．业绩越好的公司，国有股权带来的负面影响越严重。因此，基于市场承受能力的限制．从提高市场整体效率的角度出发．政府在股权分置改革第二阶段．安排受限股份上市流通时．应该考虑首先安排业绩好的上市公司的受限股份优先进入流通。实证分析结果还验证了现阶段我国市场上规模经济效应的存在。     

Inference with Dependent Data in Accounting and Finance Applications

We review developments in conducting inference for model parameters in the presence of intertemporal and cross‐sectional dependence with an emphasis on panel data applications. We review the use of heteroskedasticity and autocorrelation consistent (HAC) standard error estimators, which include the standard clustered and multiway clustered estimators, and discuss alternative sample‐splitting inference procedures, such as the Fama–Macbeth procedure, within this context. We outline pros and cons of the different procedures. We then illustrate the properties of the discussed procedures within a simulation experiment designed to mimic the type of firm‐level panel data that might be encountered in accounting and finance applications. Our conclusion, based on theoretical properties and simulation performance, is that sample‐splitting procedures with suitably chosen splits are the most likely to deliver robust inferential statements with approximately correct coverage properties in the types of large, heterogeneous panels many researchers are likely to face.     

Efficiency in the provision of social care for older people. A three-stage Data Envelopment Analysis using self-reported quality of life

Using a three-stage data envelopment analysis approach, this paper explores efficiency in the provision of social care for older people in 148 English Councils in 2009/10. Departing from D'Amico and Fernández (2012) [31] our measure of efficiency is inscribed within the production of welfare framework and based on self-reported quality of life of recipients of the services. Our results indicate a high level of efficiency, but once we control for the effects of a number of environmental variables, we found that more stringent eligibility criteria and higher assessment costs are negatively associated with the efficiency in the provision of social services.     

Does international capital flow lead to a housing boom? A time-varying evidence from China

This paper examines the causal relationship between the housing prices (HP) and the international capital flows (ICF) in China. With structural changes existing, we find that long-run relationship using full-sample data is unstable, suggesting that traditional Granger causality test is not reliable. However, we further find an unstable short-run relationship between ICF and HP when assessing the stability of the parameters and there are bidirectional causal relationships between ICF and HP for several sub-periods. Additionally, our findings indicate both positive and negative bidirectional causal relations between the series. Based on the arbitrage of ICF, the results suggest that the rise of Chinese HP is the underlying force for the inflows of international capital. Meanwhile, a surge in capital inflows may be accompanied by a rise in the price of housing. This confirms the theoretical analysis that there is an interconnected transmission mechanism between the ICF and the HP, which is diverse and depends both on the flow of ICF and on other factors.     

Is real per capita state personal income stationary? New nonlinear,asymmetric panel-data evidence

This paper re-examines the stochastic properties of U.S. state real per capita personal income, using new panel unit-root procedures. The new developments incorporate non-linearity, asymmetry, and cross-sectional correlation within panel-data estimation. Including nonlinearity and asymmetry finds that 43 states exhibit stationary real per capita personal income whereas including only nonlinearity produces 42 states that exhibit stationarity. Stated differently, we find that two states exhibit nonstationary real per capita personal income when considering nonlinearity, asymmetry, and cross-sectional dependence.     

Optimal rebalance rules for the constant proportion portfolio insurance strategy – Evidence from China

The constant proportion portfolio insurance (CPPI) strategy is one of the most popular asset allocation strategies employed by guaranteed-return financial products investors. Rebalance disciplines play an important role in determining the CPPI performance in practice. This paper examines whether the selection of rebalance rules affects CPPI strategy performance in the context of Chinese equity markets and, if so, in what pattern, and whether an optimal parameter of rebalance exists. We find that, (1) the three alternative rebalance disciplines – time discipline, market move discipline and lag discipline – are indifferent in affecting the performance of CPPI strategy; (2) in terms of optimal parameters of each rebalance rule, the optimal rebalancing period for the time discipline is 3 trading days, the optimal trading threshold of the market move discipline 4%, and the optimal lag factor of the lag discipline 6%. These optimal parameters are not influenced by the length of investment.     

The role of brand schemas,information transparency,and source of message on apparel brands' social responsibility communication

In today's complex business environment, apparel brands are communicating about their socially responsible (SR) practices through marketing messages to create a niche for themselves and show their SR efforts. However, SR aspects of products are difficult to verify by consumers. To help brands effectively communicate their SR efforts, this study assessed how consumers process an apparel brand's SR messages based on their prior experiences with that brand (brand schemas). Information transparency on messages and source of messages were tested as external cues to influence consumers' brand schemas and SR message evaluations in a mixed method repeated measures online experiment using a national US consumer sample. Results from regression-based conditional process analyses indicated that increased congruity of consumers' schemas to the fact that brands are SR led to more favorable attitude toward brands' SR messages. Also, presence of high information transparency on SR messages influenced consumers' schemas positively, and in turn, they evaluated messages more favorably. However, consumers were not influenced by whether messages were made by brands or third-party organizations. The study results show the importance of brand schemas and information transparency on apparel brand communications, helping both apparel brands and educators create effective SR-related marketing strategies.     

Efficiency Determinants and Capacity Issues in Angolan Insurance Companies

This paper describes a variety of approaches used to assess the efficiency of a sample of major insurance companies in Angola between 2003 and 2012. Starting out with the bootstrapping technique, several data envelopment analysis (DEA) estimates were generated, allowing the use of confidence intervals and bias correction in central estimates to test for significant differences in efficiency levels and input‐decreasing/output‐increasing potentials. Previous studies have focused on the measurement and explanation of the factors affecting the performance rather than the prediction. The use of neural networks combined with DEA results as part of an attempt to produce a model for insurance companies’ performance with effective predictive ability is investigated. The findings indicate that older insurance companies with Portuguese origin tend to be more efficient. Results also suggest that opportunities for accommodating future demand appear to be scarce.