Dynamic local models for segmentation and prediction of financial time series

In the analysis and prediction of many real-world time series, the assumption of stationarity is not valid. Aspecial form of non-stationarity, where the underlying generator switches between (approximately) stationary regimes, seems particularly appropriate for financial markets. We introduce a new model which combines a dynamic switching (controlled by a hidden Markov model) and a non-linear dynamical system. We show how to train this hybrid model in a maximum likelihood approach and evaluate its performance on both synthetic and financial data.     

Fast Kernel Smoothing of Point Patterns on a Large Network using Two‐dimensional Convolution

We propose a computationally efficient and statistically principled method for kernel smoothing of point pattern data on a linear network. The point locations, and the network itself, are convolved with a two‐dimensional kernel and then combined into an intensity function on the network. This can be computed rapidly using the fast Fourier transform, even on large networks and for large bandwidths, and is robust against errors in network geometry. The estimator is consistent, and its statistical efficiency is only slightly suboptimal. We discuss bias, variance, asymptotics, bandwidth selection, variance estimation, relative risk estimation and adaptive smoothing. The methods are used to analyse spatially varying frequency of traffic accidents in Western Australia and the relative risk of different types of traffic accidents in Medellín, Colombia.     

Influence of cultural divergence on consumers' perception towards reconstructed products: A qualitative enquiry

The purpose of this study is to understand whether consumers from different cultures have varied perception towards reconstructed products. Cultural divergence concept provides the theoretical support for the study. Data are collected from consumers of two culturally diverse societies (India and the United States). We use grounded theory for interview and data analysis. From the data, 23 open codes and 8 axial codes for Indian consumers and 28 open codes and 8 axial codes for the American consumers are generated. The key finding of our study suggests that culture plays a decisive role in shaping consumers' perception towards reconstructed products. The other findings of our study indicate that consumers of both societies are influenced by other factors that include product‐ or process‐related factors, brand‐ or store‐related factors, and individual factors. We discuss policy level, firm level, and nonprofit or charity level implications. Finally, limitation and scope for future research are outlined.     

AN EMPIRICAL ANALYSIS OF CHILD CARE DEMAND IN SWITZERLAND

ABSTRACT ** :  This paper analyzes the demand of Swiss families for child care facilities. A choice experiment is used to study the effects of the facilities' characteristics as well as socio-economic factors on the selected child care mode. The experimental data are analyzed using a discrete choice model with multinomial logit specification. The results suggest that the demand for extra-familial day care could be considerably higher than that observed from the actual choices constrained by insufficient provision of affordable day care. The price, access, and the quality of service as well as parents' income and education have important impacts on the choice of the mode of care.     

Female consumers as agents of change for transforming the environmental sustainability landscape

Faith in others (FIO; governments and businesses) can be seen as society's predisposition to handle environmental problems. The article examines whether women, who undertake a large majority of household purchases, can influence change in consumption practices so as to contribute to sustainability initiatives. Understanding women's character in this manner is informative, it helps us to consider how a women's role as a household purchasing agent is an important nexus to environmental sustainability. The study found that FIO has an influence on women's purchasing decisions. The study provides additional support to strengthen women's role in environmental matters. Not all women were found to be sustainability‐minded. In the challenges faced by women with respect to their influence, evidence from this research suggests that if women have more faith in governments and businesses it would make them more eco‐savvy to act as agents of change for transforming the environmental sustainability landscape.     

Fatality rate of pedestrians and fatal crash involvement rate of drivers in pedestrian crashes: a case study of Iran

The aim of this study was to uncover patterns of pedestrian crashes. In the first stage, 34,178 pedestrian-involved crashes occurred in Iran during a four-year period were grouped into homogeneous clusters using a clustering analysis. Next, some in-cluster and inter-cluster crash patterns were analysed. The clustering analysis yielded six pedestrian crash groups. Car/van/pickup crashes on rural roads as well as heavy vehicle crashes were found to be less frequent but more likely to be fatal compared to other crash clusters. In addition, after controlling for crash frequency in each cluster, it was found that the fatality rate of each pedestrian age group as well as the fatal crash involvement rate of each driver age group varies across the six clusters. Results of present study has some policy implications including, promoting pedestrian safety training sessions for heavy vehicle drivers, imposing limitations over elderly heavy vehicle drivers, reinforcing penalties toward under 19 drivers and motorcyclists. In addition, road safety campaigns in rural areas may be promoted to inform people about the higher fatality rate of pedestrians on rural roads. The crash patterns uncovered in this study might also be useful for prioritizing future pedestrian safety research areas.     

American depositary receipts: An analysis of international stock price movements

With significant increases in private capital flows across the globe, there has been a rise in the US listing of foreign stocks as American depositary receipts (ADRs). In this study, we employ cointegration techniques and estimate error-correction (EC) models to examine the degree of integration between US and three foreign equity markets. We find that ADRs are cointegrated with ordinary shares trading in the UK, Japan, and Germany, which implies that for long-term investors, they are a substitute for ordinary shares. Our analysis of the dynamic relationships between ADRs and foreign equities suggest that both markets contribute information pertinent to portfolio valuation. However, the foreign markets are found to be the more important source of information.     

Unobserved heterogeneity in stochastic cost frontier models: an application to Swiss nursing homes

Housing price-to-income and price-to-rent ratios are among the most widely monitored indicators of housing market conditions. While these ratios tend to fluctuate around a constant level or a mild trend over the long term, they also tend to deviate from these benchmarks for protracted periods. Traditional unit root tests often indicate the presence of a unit root. This article uses the framework of fractional integration to test the persistence of price-to-income and price-to-rent ratios in a sample of 16 OECD countries spanning four decades. The results indicate that the ratios are highly persistent. The possibility that persistence estimates may be affected by structural breaks in the series is also considered, but evidence of such breaks is found only in a very limited number of cases. Policy action may be required if high price-to-income and price-to-rent ratios have adverse social and economic consequences. Policies should be guided by a careful analysis of the factors behind high ratios.     

Gender-diverse boards and audit fees: What difference does gender quota legislation make?

We investigate the effect of board (audit committee) gender diversity on audit fees in the French context. We also examine whether the relationship between the proportion of female directors and audit fees is moderated by the enactment of the gender quota law in 2011. We use the system GMM estimation approach on a matched sample of French firms listed in the SBF 120 index between 2002 and 2017. Consistent with the supply-side perspective, we contend that female independent directors and female audit committee members, by improving board monitoring effectiveness, affect the auditor's assessment of audit risk, resulting in lower audit fees. Our findings also document that, by breaking the glass ceiling, the effectiveness of the gender quota law lies not in increasing the proportion of female insider directors, but in boosting the appointment of female independent directors and female audit committee members. Using the difference-in-difference approach, our results reveal that female independent directors and female audit committee members are more willing to assert their monitoring skills after the quota law, leading to lower audit fees. Moving beyond tokenism, we show that, after the quota law, the negative impact on non-audit fees is strengthened only for female independent directors.