Nearly exact sample size calculation for powerful non‐randomized tests for differences between binomial proportions

In the case of two independent samples, it turns out that among the procedures taken in consideration, BOSCHLOO'S technique of raising the nominal level in the standard conditional test as far as admissible performs best in terms of power against almost all alternatives. The computational burden entailed in exact sample size calculation is comparatively modest for both the uniformly most powerful unbiased randomized and the conservative non‐randomized version of the exact Fisher‐type test. Computing these values yields a pair of bounds enclosing the exact sample size required for the Boschloo test, and it seems reasonable to replace the exact value with the middle of the corresponding interval. Comparisons between these mid‐N estimates and the fully exact sample sizes lead to the conclusion that the extra computational effort required for obtaining the latter is mostly dispensable. This holds also true in the case of paired binary data (McNemar setting). In the latter, the level‐corrected score test turns out to be almost as powerful as the randomized uniformly most powerful unbiased test and should be preferred to the McNemar–Boschloo test. The mid‐N rule provides a fairly tight upper bound to the exact sample size for the score test for paired proportions.     

An intelligent service matching method for mechanical equipment condition monitoring using the fibre Bragg grating sensor network

Due to the advantages of being able to function under harsh environmental conditions and serving as a distributed condition information source in a networked monitoring system, the fibre Bragg grating (FBG) sensor network has attracted considerable attention for equipment online condition monitoring. To provide an overall conditional view of the mechanical equipment operation, a networked service-oriented condition monitoring framework based on FBG sensing is proposed, together with an intelligent matching method for supporting monitoring service management. In the novel framework, three classes of progressive service matching approaches, including service-chain knowledge database service matching, multi-objective constrained service matching and workflow-driven human-interactive service matching, are developed and integrated with an enhanced particle swarm optimisation (PSO) algorithm as well as a workflow-driven mechanism. Moreover, the manufacturing domain ontology, FBG sensor network structure and monitoring object are considered to facilitate the automatic matching of condition monitoring services to overcome the limitations of traditional service processing methods. The experimental results demonstrate that FBG monitoring services can be selected intelligently, and the developed condition monitoring system can be re-built rapidly as new equipment joins the framework. The effectiveness of the service matching method is also verified by implementing a prototype system together with its performance analysis.     

Gender diversity,gender norms and firm performance: Evidence from India

This paper examines the effect of gender diversity on the financial performance of firms in India. Specifically, we examine two specific aspects of gender diversity- how the implementation of gender diversity across various workforce levels affects firm performance and the role of gender norms and culture in the external environment where the firm operates in influencing the relation. Using the World Bank Enterprise Survey, 2014, we use propensity score matching and instrumental variable estimation techniques to test the effect of gender diversity on firm performance. The results suggest that better women’s representation at higher levels of the workforce is associated with better performance. In contrast, women’s representation at the worker level of the workforce does not affect performance. Additionally, the state-level gender empowerment norms strongly condition the effect of gender diversity on performance. In states with better gender empowerment indicators, firms with better female representation at all levels of permanent employment also have better performance. In contrast, there is no effect of gender diversity on performance in states with worse women empowerment indicators.     

Testing the null hypothesis of no regime switching with an application to GDP growth rates

This paper presents tests for the null hypothesis of no regime switching in Hamilton’s (Econometrica 57:357–384, 1989) regime switching model. The test procedures exploit similarities between regime switching models, autoregressions with measurement errors, and finite mixture models. The proposed tests are computationally simple and, contrary to likelihood based tests, have a standard distribution under the null. When the methodology is applied to US GDP growth rates, no strong evidence of regime switching is found. I thank Don Andrews, Peter Phillips, Yuichi Kitamura, Anat Bracha, Patrik Guggenberger, Orit Whiteman and three anonymous referees for useful comments and suggestions.     

Choosing propensity score matching over regression adjustment for causal inference: when,why and how it makes sense

Summary

This study identified when regression adjustment fails to adjust adequately for differences in observed covariates and where propensity score matching is the only alternative.

Multivariate analysis might fail to adjust for observed confounders if:

• 1. The means of the propensity scores in the two groups are more than one-half a standard deviation apart unless distributions of the covariates in both groups are nearly symmetric, sample sizes of the two groups are approximately the same and distributions of the covariates in the two groups have similar variances;

• 2. The ratio of the propensity score variances in the two groups is significantly different from one;

• 3. The ratio of residual variances in the two groups after adjusting for the propensity score is significantly different from one.

Conducted retrospective analysis showed that the treatment effect would be an estimated $305 (or 26%) less if the misspecified outcome model had been chosen.     

Entrepreneurial innovativeness and growth ambitions in thick vs. thin regional innovation systems

Research in economic geography has paid increasing attention to regional innovation systems (RISs) as a potential vehicle for growth and development. Yet despite an increasing amount of research studying RISs in particular and economic regions in general, we have limited knowledge about their influence on entrepreneurs and entrepreneurship. We respond to this knowledge gap and study if entrepreneurs’ localization in thick vs. thin RISs affects their innovativeness and growth ambitions. Thick RISs are predominately urbanized spaces that include organizations of higher-level education, R&D intensive milieus, and an ample industry sector, while thin RISs to a lesser degree encompass these features. Empirically, we analyse 870–917 entrepreneurial firms in Agder of Southern Norway. Based on trade and labour markets, as defined by the EU’s classification of local administrative units (LAU1), we identify two thick and six thin RISs in Agder. Econometric analyses show that entrepreneurs located in thick RISs are more innovative than entrepreneurs located in thin RISs, but there are no significant differences concerning entrepreneurs’ growth ambitions. In light of our findings, we discuss the potential agency role played by entrepreneurial firms at a micro level on path dependent features of RISs at a macro level.     

Model uncertainty and asset return predictability: an application of Bayesian model averaging

We investigate model uncertainty associated with predictive regressions employed in asset return forecasting research. We use simple combination and Bayesian model averaging (BMA) techniques to compare the performance of these forecasting approaches in short-vs. long-run horizons of S&P500 monthly excess returns. Simple averaging involves an equally-weighted averaging of the forecasts from alternative combinations of factors used in the predictive regressions, whereas BMA involves computing the predictive probability that each model is the true model and uses these predictive probabilities as weights in combing the forecasts from different models. From a given set of multiple factors, we evaluate all possible pricing models to the extent, which they describe the data as dictated by the posterior model probabilities. We find that, while simple averaging compares quite favorably to forecasts derived from a random walk model with drift (using a 10-year out-of-sample iterative period), BMA outperforms simple averaging in longer compared to shorter forecast horizons. Moreover, we find further evidence of the latter when the predictive Bayesian model includes shorter, rather than longer lags of the predictive factors. An interesting outcome of this study tends to illustrate the power of BMA in suppressing model uncertainty through model as well as parameter shrinkage, especially when applied to longer predictive horizons.     

食品安全规制能提高生产者福利吗？——基于不同规制强度的检验

食品安全规制涉及政府、生产者和消费者三方利益主体的博弈。已有文献主要关注政府规制对减少生产者违规概率、提升食品安全水平和消费者福利所产生的作用。本文从生产者福利视角考察企业参与食品安全规制是否存在“收益率溢价”效应,以此对中国食品安全规制的间接效果进行实证检验。首先采用倾向得分匹配模型考察食品安全规制对企业资产收益率的平均影响效应,然后采用广义倾向得分匹配模型和门槛模型进一步考察不同规制强度与企业资产收益率的非线性关系。研究表明,当前食品安全规制对参与规制企业产生了“收益率溢价”的间接效果,但随着规制强度的扩展,企业资产收益率呈现先上升后下降的“倒U型”变化,具有显著的双重门槛特征,当企业持有4项质量认证证书时为“最优规制强度”。最后结合当前中国实际,解释了上述结论产生的原因及政策含义。     

The impact of minimum wages on employment of low‐wage workers

This study provides empirical evidence on the impact of a minimum wage increase on employment of workers in the formal sector who have wages below the minimum level in Vietnam. Using the difference‐in‐differences with propensity score matching and the Vietnam Household Living Standard Surveys of 2004 and 2006, the article finds that the minimum wage increase in 2005 reduced the proportion of workers having a formal sector job among low‐wage workers. Most workers who lost formal sector jobs became self‐employed.     

Schumpeter,Schumacher and the Greening of Technology

We propose a semantic patent claim analysis that can examine patents for possible infringements and identify which needs to be manually perused. So far, numerous approaches have been devised to systemise this burden, but have not been useful in practice because of a lack of consideration of semi-structure of patent claim data and claim element-based procedure of adjudicating patent infringement. At the heart of our method is a hierarchical keyword vector for representing the dependency relationships among claim elements (as well as unstructured textual information) and a tree matching algorithm for comparing claim elements of patents. A case study of the patents about DNA chip technology shows our method has considerable advantages in terms of accuracy and significance. We believe the suggested method could be employed in various research areas and serve as a starting point for developing more general models.