Spike-and-slab type variable selection in the Cox proportional hazards model for high-dimensional features

In this paper, we develop a variable selection framework with the spike-and-slab prior distribution via the hazard function of the Cox model. Specifically, we consider the transformation of the score and information functions for the partial likelihood function evaluated at the given data from the parameter space into the space generated by the logarithm of the hazard ratio. Thereby, we reduce the nonlinear complexity of the estimation equation for the Cox model and allow the utilization of a wider variety of stable variable selection methods. Then, we use a stochastic variable search Gibbs sampling approach via the spike-and-slab prior distribution to obtain the sparsity structure of the covariates associated with the survival outcome. Additionally, we conduct numerical simulations to evaluate the finite-sample performance of our proposed method. Finally, we apply this novel framework on lung adenocarcinoma data to find important genes associated with decreased survival in subjects with the disease.     

Nonprofit fundraising with virtual reality

An increasingly competitive fundraising environment and the new media ecosystem have prompted nonprofit organizations to strengthen their digital marketing capabilities. Recently, a few nonprofits have used virtual reality (VR) technology in fundraising, which raises questions about its effectiveness. Does VR work, and if so, through what psychological mechanisms? Through a lab experiment, the usefulness of VR as a digital fundraising tool was investigated. Specifically, we compared the media effects of experiencing the same fundraising video on two different devices: (a) a head‐mounted VR and (b) a tablet. The results revealed that donation intention, perceived vividness, perceived interactivity, and social presence were all significantly greater with the VR medium than with the tablet medium. Next, we tested and successfully verified the mediation effect of social presence on donation intention. Finally, we verified that a viewer's sensation‐seeking tendency served as a moderator when the device type influenced donation intention. Specifically, the media effects of VR were stronger for high sensation seekers than for low sensation seekers. This study contributes to theory and research by verifying the VR fundraising effect and identifying the key role of social presence, and it also provides important managerial guidelines for media design in VR fundraising campaigns.     

Pricing Policy in a Supply Chain: Negotiation or Posted Pricing

This article examines the choice of pricing policy (posted pricing or negotiation) toward end customers in a supply chain. Many retailers actively decide whether or not to encourage negotiation on the shop floor. Of course, the retailer's pricing policy influences not only the retailer's profit, but also the profits of the manufacturers who sell through the retailer. However, little is known about the forces that shape the pricing policy when two self‐interested parties interact in a supply chain. We consider two alternative models depending on who has the power to decide the pricing policy: the manufacturer or the retailer. We find that an increase in the wholesale price weakens the retailer's ability to price discriminate through negotiation. Therefore, the retailer prefers negotiation at lower wholesale prices and posted pricing at higher wholesale prices. We also find that whenever the retailer prefers negotiation, the manufacturer does too. Therefore, the retailer's discretion over the pricing policy causes friction only when the retailer wants to use posted pricing, while the manufacturer wishes the retailer to use negotiation. We show that such friction arises only when product availability or the cost of negotiation is moderate. In this case, we show that the manufacturer may offer a substantial discount to persuade the retailer to negotiate. Surprisingly, in this region of friction, a decrease in the supply chain's capacity or an increase in negotiation costs (both of which are typically considered as worsening the retailer's business environment) translates into higher profit for the retailer.     

Analysis of fertility using cohort-specific socio-economic data

Review of Economics of the Household - In this paper we propose a novel approach that relates age-specific cohort fertility rates with socioeconomic conditions faced by each cohort in each single...     

LOCUS OF CONTROL AND JOB TURNOVER

In this article, I investigate how a worker's locus of control, that is, the perception of control over daily events, affects job‐to‐job and job‐to‐nonemployment turnover. I find that an increase in internality—the degree to which respondents believe that the outcomes of their life events are determined by their own actions versus external factors—increases job‐to‐job transitions. In addition, the annual wage growth rate and the wage gains from job‐to‐job transitions increase with internality. The influence of the locus of control on job‐to‐nonemployment turnover, however, is insignificant on controlling for the worker's level of attained education. (JEL D83, J63)     

Risk analysis models and risk degree determination in new product development: A case study

This paper proposes a risk analysis model to determine the risk degrees of the risk factors occurring in product development processes. The model uses both fuzzy theory and Markov processes on a concurrent engineering (CE) basis. Fuzzy models determine the impact values of the risk factors, and Markov processes determine the probability of risk occurrences. The analysis model is used to compute the risk degrees by multiplying the probability of risk occurrences by the impact value. This study can be utilized for analyzing the influences of risk factors on product development projects and will contribute toward the development of a risk management framework (RMF) to defend against various risk factors. Implications and directions for future research are discussed.     

Impacts of buyers' order batching on the supplier's demand correlation and capacity utilization in a branching supply chain

There are several ways for a manufacturer to cope with demand uncertainty, e.g. inventories, capacity and cash. Among these, this study focuses on the second one, the capacity, especially on the problem of investing in flexible facilities and enhancing their utilization via demand management. In a supply chain, demands that an upstream firm (supplier) faces are the purchase orders from the downstream members (buyers). We analyse the impacts of buyers' order batching on the supplier's demand correlation and capacity utilization in a simple branching supply chain, where a supplier does business with two buyers whose market demands are correlated. Our results show that: (i) a supplier whofacesa smaller demand correlation coefficient (i.e. closer to-1) would invest more in flexible facilities; (ii) an increase in order lot size mitigates the correlation of purchase orders; and (iii) a supplier whose facilities are flexible would prefer frequent orders with smaller lots only when market demands are highly negatively correlated. This means that even suppliers whose facilities are flexible would rather prefer infrequent orders with larger lots in the presence of positively correlated demands. Additionally, some managerial implications are discussed.     

Explaining Korean Welfare State Development with New Empirical Data and Methods

Selecting Korea as one of the conceptually classified East Asian welfare states, this paper aims to empirically investigate the determinants of Korean welfare state development. Linking the research question with the lively discussion on welfare state in Korea that is taking place domestically, this paper also aims to examine whether the development of the Korean welfare state has been influenced by politics of the left or by industrialization with functionalistic perspectives. We conduct an analysis first by regression analysis and then fuzzy‐set qualitative comparative analysis (fsQCA) to examine the conditions of such development. A Korean Welfare State Dataset (KWSD) has been newly included in the Comparative Welfare State Dataset (CWSD) and we used this new dataset for our empirical analysis. Our empirical findings suggest that, at least up to the current point, factors other than economic development or industrialization are not robust enough to make a claim of a take‐off of the Korean welfare state from its developmental stage. However, we also notice the influence of the left government and suggest that it is legitimate to expect an innovative change in the path of the Korean welfare state once the power resource theory is applicable.     

The effects of adoption disclosure on the adaptation of Korean adoptive families

The main goal of this study is to examine the effects of adoption disclosure on the adaptation of Korean adoptive families using a two stage equation model. The results of this study indicate that there is a self-selection bias between disclosed and closed adoptive families. Yet, even after considering this selection bias, adoption disclosure was associated with the positive outcomes of adoptive families. Also, the estimates in the treatment effect model were found to be very similar to those of the regression model, indicating that the degree of self-selection bias between these groups was not so strong as to distort the effects of adoption disclosure and other predictors.