Risk spillover between Bitcoin and conventional financial markets: An expectile-based approach

In order to challenge the existing literature that points to the detachment of Bitcoin from the global financial system, we use daily data from August 17, 2011–February 14, 2020 and apply a risk spillover approach based on expectiles. Results show reasonable evidence to imply the existence of downside risk spillover between Bitcoin and four assets (equities, bonds, currencies, and commodities), which seems to be time dependent. Our main findings have implications for participants in both the Bitcoin and traditional financial markets for the sake of asset allocation, and risk management. For policy makers, the findings suggest that Bitcoin should be monitored carefully for the sake of financial stability.     

Predictive regressions under asymmetric loss: Factor augmentation and model selection

This paper discusses the specifics of forecasting using factor-augmented predictive regressions under general loss functions. In line with the literature, we employ principal component analysis to extract factors from the set of predictors. In addition, we also extract information on the volatility of the series to be predicted, since the volatility is forecast-relevant under non-quadratic loss functions. We ensure asymptotic unbiasedness of the forecasts under the relevant loss by estimating the predictive regression through the minimization of the in-sample average loss. Finally, we select the most promising predictors for the series to be forecast by employing an information criterion that is tailored to the relevant loss. Using a large monthly data set for the US economy, we assess the proposed adjustments in a pseudo out-of-sample forecasting exercise for various variables. As expected, the use of estimation under the relevant loss is found to be effective. Using an additional volatility proxy as the predictor and conducting model selection that is tailored to the relevant loss function enhances the forecast performance significantly.     

Automated topic modeling of tourist reviews: Does the Anna Karenina principle apply?

Automated content analysis of online travel reviews allows identification of topics of travelers' satisfaction, yet its domain is not well researched. We suggest that the Anna Karenina principle positing a greater variability of the factors leading to business failure as opposed to those leading to success can be applied to the domain of visitors’ reviews of historic and cultural attractions. The larger variability of issues in reviews of dissatisfied visitors is likely to result in limitations for automated topic modeling. We confirm our proposition using TripAdvisor reviews of the Terracotta Army museum in China, and validate the outcome with two additional sites. The study strongly suggests that application of unsupervised topic mining algorithms to negative reviews may be problematic and the results should be treated with caution. The main themes of dissatisfaction of visitors to all three sites are reported and practical implications for management of the attractions are discussed.     

Expenditure on different categories of meat in Greece: the influence of changing tastes

This paper employs a latent variable approach to isolate the effects of changing tastes on the share of total meat expenditure on different categories of meat products in Greece during the period 1965–1995. We find that changes in the relative expenditure on different categories of meat cannot be explained by changes in the relative prices of the different meat products and increased expenditure alone. For pork products in particular, the increase in the share of expenditure has been greater than would be expected as a result of the relative fall in their price. The increase can therefore be associated with changes in taste. This finding is of general interest to those conducting empirical research into consumer behaviour both in economies where there have been significant changes in patterns of food consumption, and where, as in the case of many less industrialised economies, rapid structural changes in food consumption patterns are still to come. It is also of importance to policy makers in assessing die effectiveness of advertising or promotional campaigns in influencing longer term changes in consumer preferences for different products.     

Applying LSA text mining technique in envisioning social impacts of emerging technologies: The case of drone technology

This research proposes a novel method of identifying and understanding the holistic overview of emerging technologies’ unintended consequences. Latent Semantic Analysis (LSA) text mining technique is employed to yield multiple groups of contextually similar terms from future-oriented data sources, comprising both experts’ and the public's concerns regarding future technologies. Resulting term clusters are considered as new abstractions, or so-called scenarios, of future social impacts. Furthermore, the study acquires greater depth and breadth in conceptualizing social impacts through considering condition- and value-related terms as key linking factors to previous social impact-related literature. Our proposed methodology seeks to gain insights into the utilization of future-oriented data sources for the foresight activity, hoping to mitigate public skepticism and pursue a better social acceptance of emerging technologies.     

Application of structural equation modelling for estimating airline performance

Structural equation modelling (SEM) is an estimation method that can handle a large number of exogenous and endogenous factors as well as unobserved (latent) variables that are specified as linear combinations of observed (measurement) factors. This paper presents a SEM procedure with latent variables for estimating the financial and non-financial performance in airline companies. The model includes independent, mediator and dependent latent variables. The sample for this research comprises 214 airline companies. The results indicate that the model is capable of estimating performance with respect to the economic situation and it has also been determined that two types of constructs affect performance. One type is the economic situation acting as an initial construct and the other is the internal operation acting as the mediator in the research model.     

专题微课生物化学教学新模式的探索与实践

专题微课教学是基于一门学科或课程的某个重要专题设计开发的教学模式。课题组在生物化学的课程中设计了专题微课。生物化学的专题微课教学包括确立教学目标、教学方法、微课录制与编辑加工等环节，应注意建立全新的网络教学平台、互动式教学、抓好指导小组资料整理等教学策略，其成功需要教师有较高的综合素质和教学技能。实践证明，生物化学的专题微课教学取得了良好的教学效果，学生的成绩有了一定提高。     

ALTERNATIVE WEIGHTING APPROACHES TO COMPUTING INDEXES OF ECONOMIC ACTIVITY

This paper compares the weighting schemes in the traditional, principal component and dynamic factor approaches to summarizing information from a number of component variables. To facilitate the comparison, we propose a framework to discuss the approaches with respect to their implied loadings in a latent variable model. We also propose a way to transform the dynamic factor index into an analogous index which is a weighted average of the components. The framework shows the strengths and weakness of the alternative weighting schemes and the sense in which the dynamic factor approach has the advantage of capturing both the significance and the variability of the components.     

The financial advice puzzle: The role of consumer heterogeneity in the advisor choice

We investigate motivational drivers of financial advisor use, accounting for investor heterogeneity, with the goal of helping institutions increase the use of financial advisor services. The results from a latent class choice model reveal two distinct segments that differ in their approach to the financial advice decision. While higher levels of risk tolerance, trust, and self-efficacy increase financial advice use for both segments, albeit at much higher propensities for Segment 1, personality only matters for Segment 1. Moreover, their regulatory focus differs with Segment 1 being promotion and Segment 2 being prevention focused. Using these results, we offer suggestions for marketing strategies.     

e-banking culture: A comparison of EU 27 countries and Portuguese case in the EU 27 retail banking context

The aim of this research is two-fold. Firstly we conducted a wise comparative analysis of EU 27 countries and secondly we identified distinct e-banking user segments of Portuguese citizens. We used a questionnaire for data collection (n=2358) and Latent Class Models (LCM) for data analysis. LCM revealed three segments in EU 27 (Portugal is in the poorest cluster, concerning the indicators used, with Greece, Spain and Cyprus) and two e-banking user segments in Portuguese citizens: Those who do not risk (49%), and Those who risk (51%). Our findings enable e-banking managers for taking appropriate strategic decisions.