Sample Selection Bias Research Articles

Economic policy uncertainty and corporate leverage - does cash holdings matter? Evidence from the U.S.

This research study examines the mediating role of cash holdings between the economic policy uncertainty (EPU) and corporate leverage relationship. Using stepwise regression analysis and annual firm-level data of 2,534 U.S. firms listed at NYSE over 1995–2018, we provide novel evidence that cash holdings significantly and partially mediate the EPU-leverage relationship, accounting for a 10.72% increase in the corporate leverage during EPU. We discover that the mediating role of cash holdings between EPU and leverage is sensitive to firm-level heterogeneity. Also, the mediating effect of cash holdings remains significant based on long- and short-term leverage. Finally, our findings are robust to outliers’ effect, alternate EPU measurement, endogeneity concerns and sample-selection bias. The findings of this study highlighted the role of firm cash holdings in a firm’s leverage decision during economic uncertainty and recommends increasing debt financing to incentivize value-increasing decisions and mitigate agency problems by reducing the cost of free cash flows. Policymakers should be aware of public and private firms’ challenges during high EPU. As we showed in this study, firms have to adjust their financial decisions during high economic uncertainty. Also, the study suggests that policymakers should maintain economic policies to avoid external shocks that may force firms to adjust their financial decision-making.

Decorrelated spectral regression: An unsupervised dimension reduction method under data selection bias

In manipulating high-dimensional data arising from real applications, dimension reduction methods have been proven to be effective and necessary. Nevertheless, the widespread selection bias, induced in the process of data collection, has long been unaware and its negative effects have been ignored. The accidental spurious correlation between the features brought by selection bias will deteriorate the performance of dimension reduction approaches. It is vital to remove such bias and consider its influence on dimension reduction. In this paper, we propose a novel decorrelated spectral regression algorithm for unsupervised dimension reduction on data with sample selection bias. In this method, the sample weights of the global feature distribution are learned by introducing the decorrelation regularizer. The learned weights are then combined with the regression model to obtain the dimension reduction mapping. Thus, the dimension reduction results are less affected by the accidental spurious correlation of data caused by selection bias. In addition, we also propose the updating rules of the parameters in our algorithm. A large number of experimental results on real-world datasets show that our method has achieved a significant improvement, indicating that it is effective and necessary to remove the accidental spurious correlation between features caused by selection bias in dimension reduction.

Self-reported reasons for having difficulty reaching orgasm in men with diverse etiologies

Abstract Background Difficulty reaching orgasm/ejaculation during partnered sex, a primary characteristic of delayed or absent ejaculation, affects about 5% to 10% of men, but the reasons underlying this problem are poorly understood. Aim The study sought to gain insight into possible etiologies of delayed ejaculation by assessing men’s self-perceptions as to why they experience difficulty reaching orgasm. Methods We drew 351 men reporting moderately severe to severe difficulty reaching orgasm during partnered sex from a sample of over 3000 respondents obtained through an online survey. As part of the 55-item survey, participants responded to 2 questions asking about their self-perceived reasons for having difficulty reaching orgasm and selected from a list of 14 options derived from the research literature, a series of men’s focus groups, and expert opinion. The first question allowed respondents to select all the reasons that they felt contributed to the problem, the second to select only the most important reason. In addition, both men with and without comorbid erectile dysfunction were investigated and compared. Outcomes Hierarchical ordering of men’s self-pereceived reasons for having difficulty reaching orgasm, including typal reasons established through principal component analysis. Results The major reasons for difficulty were related to anxiety/distress and lack of adequate stimulation, with relationship and other factors endorsed with lower frequency. Further exploration using principal components analysis identified 5 typal reasons, in descending order of frequency: anxiety/distress (41%), inadequate stimulation (23%), low arousal (18%), medical issues (9%), and partner issues (8%). Few differences emerged between men with and without comorbid ED other than ones related to erectile problems, such as higher level of endorsement of medical issues. Typal reasons showed correlations, albeit mostly weak, with a number of covariates, including sexual relationship satisfaction, frequency of partnered sex, and frequency of masturbation. Clinical Implications Until supplemental medical treatments for delayed ejaculation are developed and approved, a number of men’s purported reasons for difficult or absent ejaculation/orgasm—anxiety/distress, inadequate stimulation, low arousal, relationship issues—fall into areas that can be addressed in couples counseling by a trained sex therapist. Strengths and Limitations This study is unique in scope and robust in sample size. Drawbacks include those associated with online surveys, including possible bias in sample selection, limitation to Western-based samples, and the lack of differentiation between men with lifelong and acquired difficulty. Conclusion Men who have difficulty reaching ejaculation/orgasm identify putative reasons for their problem, ranging from anxiety/stress, inadequate stimulation, and low arousal to partner issues and medical reasons.

PatchMix Augmentation to Identify Causal Features in Few-shot Learning.

The task of Few-shot learning (FSL) aims to transfer the knowledge learned from base categories with sufficient labelled data to novel categories with scarce known information. It is currently an important research question and has great practical values in the real-world applications. Despite extensive previous efforts are made on few-shot learning tasks, we emphasize that most existing methods did not take into account the distributional shift caused by sample selection bias in the FSL scenario. Such a selection bias can induce spurious correlation between the semantic causal features, that are causally and semantically related to the class label, and the other non-causal features. Critically, the former ones should be invariant across changes in distributions, highly related to the classes of interest, and thus well generalizable to novel classes, while the latter ones are not stable to changes in the distribution. To resolve this problem, we propose a novel data augmentation strategy dubbed as PatchMix that can break this spurious dependency by replacing the patch-level information and supervision of the query images with random gallery images from different classes from the query ones. We theoretically show that such an augmentation mechanism, different from existing ones, is able to identify the causal features. To further make these features to be discriminative enough for classification, we propose Correlation-guided Reconstruction (CGR) and Hardness-Aware module for instance discrimination and easier discrimination between similar classes. Moreover, such a framework can be adapted to the unsupervised FSL scenario. The utility of our method is demonstrated on the state-of-the-art results consistently achieved on several benchmarks including miniImageNet, tieredImageNet, CIFAR-FS , CUB, Cars, Places and Plantae, in all settings of single-domain, cross-domain and unsupervised FSL. By studying the intra-variance property of learned features and visualizing the learned features, we further quantitatively and qualitatively show that such a promising result is due to the effectiveness in learning causal features.

Adjusting for Selection Bias in Nonprobability Samples by Empirical Likelihood Approach

Abstract Large amount of data are today available, that are easier and faster to collect than survey data, bringing new challenges. One of them is the nonprobability nature of these big data that may not represent the target population properly and hence result in highly biased estimators. In this article two approaches for dealing with selection bias when the selection process is nonignorable are discussed. The first one, based on the empirical likelihood, does not require parametric specification of the population model but the probability of being in the nonprobability sample needed to be modeled. Auxiliary information known for the population or estimable from a probability sample can be incorporated as calibration constraints, thus enhancing the precision of the estimators. The second one is a mixed approach based on mass imputation and propensity score adjustment requiring that the big data membership is known throughout a probability sample. Finally, two simulation experiments and an application to income data are performed to evaluate the performance of the proposed estimators in terms of robustness and efficiency.

Correcting sample selection bias with model averaging for consumer demand forecasting

Correcting sample selection bias with model averaging for consumer demand forecasting

Are real interest rates a monetary phenomenon? Evidence from 700 years of data

In this paper we examine the effect of permanent inflation shocks on real interest rates, based on a structural Time-Varying Parameter Vector Autoregression (TVP-VAR) model that accounts for parameter instability, using the most extensive annual dataset that accounts for the entire economic history that dates back to 1310 for France, Germany, Holland (the Netherlands), Italy, Japan, Spain, the United Kingdom (U.K.) and the United States (U.S.). The Fisherian hypothesis of a one-to-one movement of inflation to nominal interest rates can only be rejected episodically, in favour of a Mundell-Tobin effect of less than proportional increase in the nominal interest rate to an inflation shock. Our findings suggest that long-run real interest rates of advanced economies have historically remain unaffected by inflation shocks due to a corresponding one-to-one increase in the nominal interest rate. Hence, the conclusions drawn by the majority of the existing literature based on post World War II samples should be treated with caution, due to sample selection bias.

Can contract farming improve farmers’ technical efficiency and income? Evidence from beef cattle farmers in China

IntroductionContract farming is seen as a tool to create new market opportunities that can address market imperfections in many developing countries and thus increase smallholder income.MethodsThis study examines the impact of contract farming on farm household income using survey data from 610 rural households in China. The propensity score matching method addresses the sample selection bias of participation in contract farming.ResultsContract farming can significantly increase farmers’ income, and both marketing contracts and production-management contracts can substantially increase farmers’ income levels, with production-management contracts having a greater degree of impact. Additional analysis reveals that breeding years, farm size, and training time can significantly affect how contract farming enhances farmers’ income. At the same time, contract farming can also considerably improve farmers’ technical efficiency in agricultural production. Participation in contract farming enhances the tendency to centralize the technical efficiency of agricultural production. Further analysis shows that the technical efficiency of agricultural production partially mediates the effects of contract farming on farm household income.DiscussionContract farming can be an effective institutional arrangement for improving the technical efficiency of farm household production and revenue. We also point out that farmers should strengthen their comprehensive ability levels and actively participate in training to acquire new knowledge and improve their cognitive ability. Simultaneously, according to the characteristics of farmers’ resource endowments, small farmers are encouraged to cooperate with companies in depth and develop contractual contracts in a targeted manner. Promote win-win cooperation and benefit-sharing among various business entities to promote the sustainable and high-quality development of China’s beef cattle industry.

Social identity of civil servants and online peer-to-peer lending: Evidence from China

Using a novel dataset of 1,777,847 peer-to-peer (P2P) transactions between 2010 and 2017, we investigate how the social identity of Chinese civil servants affects P2P lending. The empirical results indicate that the social identity of civil servants leads to a higher probability of obtaining loans and a lower interest rate in P2P lending. Our results thus reveal that Chinese civil servants obtain considerable benefits in the P2P lending market compared with non-civil servants; these empirical results remain robust after considering sample selection bias.

Solar PV adoption in rural Burkina Faso

Energy is a rare commodity in Africa. Many households use unclean fuels for domestic activities. While more than 90% of rural households use fuelwood and kerosene as a source of energy in Sub-Saharan Africa, this study examines the determinants of energy diversity through solar PV adoption by rural household. Employing primary data on 105 villages from Burkina Faso, a sample of 6300 households is investigated. Performing the probit and using a sample selection bias correction technique, the findings show that rural households engaged in economic activities are more likely to adopt a solar PV system. The entrepreneurial spirit has a positive impact on the adoption of solar PV. Moreover, the use of solar PV technology is strongly associated with household head age, gender, family size, and education. The findings suggest that financial support and micro-credit must target more rural households.

Green Credit Policy, Environmental Investment, and Green Innovation: Quasi-Natural Experimental Evidence from China

In order to explore whether green credit policy can guide the green transformation of heavily polluting firms, we examine the influence of green credit policy on green innovation. Further, we analyze the mediating effect of environmental investment and the moderating effect of type of ownership and green finance development level in this relationship. Findings from the DID model indicate that the Green Credit Guidelines led to a significant increase in green innovation at heavily polluting enterprises, both quantitatively and qualitatively, with environmental investment acting as partial mediators. Further, the positive influence of green credit policy is more substantial in state-owned firms and in regions with high levels of green finance development. Findings are robust and remain valid after different sensitivity tests, including the improved PSM-DID model and the elimination of interference from some samples to address the sample selection bias existing in the DID model.

Has ESG Performance Reduced Stock Price Volatility

In the context of sustainable development, it is of practical importance to study whether ESG performance can influence stock price volatility. This paper empirically analyzes the impact of corporate ESG performance on stock price volatility using a sample of A-share listed companies in Shanghai and Shenzhen, China, from 2011-2021. It is found that corporate ESG performance can significantly reduce stock price volatility, and this finding remains robust after controlling for sample selection bias, changing the measurement of key variables, and transforming the empirical model. Mechanistic tests suggest that analyst attention and corporate reputation are potential mechanisms of influence between ESG performance and stock price volatility. Further study finds that investor sentiment contributes to ESG performance in reducing stock price volatility. This paper provides operational evidence for ESG performance to play a positive screening role, stabilize stock markets and promote high-quality development of listed companies.

The social value of a SARS-CoV-2 vaccine: Willingness to pay estimates from four western countries.

SARS-CoV-2 vaccines give rise to positive externalities on population health, society and the economy in addition to protecting the health of vaccinated individuals. Hence, the social value of such a vaccine exceeds its market value. This paper estimates the willingness to pay (WTP) for a hypothetical SARS-CoV-2 vaccine (or shadow prices), in four countries, namely the United States (US), the United Kingdom, Spain and Italy during the first wave of the pandemic when COVID-19 vaccines were in development but not yet approved. WTP estimates are elicited using a payment card method to avoid "yea saying" biases, and we study the effect of protest responses, sample selection bias, as well as the influence of trust in government and risk exposure when estimating the WTP. Our estimates suggest evidence of an average value of a hypothetical vaccine of 100-200 US dollars once adjusted for purchasing power parity. Estimates are robust to a number of checks.

Understanding the characteristics of car-sharing users and what influences their usage frequency

Throughout the global market, the car-sharing industry continues to confront multiple headwinds. In China, the problem of the generally low frequency of car-sharing use by users (FCU) is a major constraint to its growth. In this study, we quantify the characteristics of car-sharing users and the main factors influencing the FCU based on the car-sharing order data and GPS trajectory data of users in Beijing, China. Considering that sample selection bias with car-sharing users can lead to biased and inconsistent parameter estimation, this paper constructs a multi-factor influence model of FCU based on the Heckman two-stage model. Particularly, whether users enter the market through coupons (IF-Coupons) is selected as the instrumental variable of the model. The results reveal that current car-sharing users are mainly young people aged 25–39 and predominantly men. Travelers tend to travel on weekdays, and those making short trips, and those who enter the car-sharing market using coupons are more likely to become car-sharing users. Based on the results for the FCU, the market potential of female car-sharing users is larger. Besides, the indicators of travel distance and duration, the tendency to use on weekdays, and station location can significantly influence the FCU. Our research contributes to provide scientific theoretical support for promoting the coordinated and sustainable development of the car-sharing market.

Farmers’ Endowments, Technology Perception and Green Production Technology Adoption Behavior

The motivation of this paper is to explore the influence of farmers’ endowments and technology perceptions on farmers’ green production technology adoption behaviors. Based on a survey of 471 grain farmers in the main grain-producing areas of Shandong Province, the Heckman model was used to empirically analyze the influence of farmers’ endowments on their adoption behavior of green production technologies, and the moderating effect of technology perception on the relationship between the two was further investigated. The study showed that: (1) More than half of the farmers adopted more than three green production technologies, and only 11.5% did not adopt them. In terms of adoption structure, the adoption rate of green pest control technology was the highest at 67.7%, while the adoption rate of moderate deep pine technology was the lowest at 32.7%. The adoption structure should be further optimized. (2) After controlling for sample selection bias, farmers’ endowments have a significant positive effect on farmers’ green production technology adoption behaviors. The ordinary least square regression overestimates the main effect of farmers’ endowments by 8.5% compared with the Heckman model. (3) Technology cognition can effectively promote the positive effect of farmers’ endowments on green production technology adoption, in which the moderating effect of technology–economic cognition is higher than that of technology–environmental cognition. (4) There is heterogeneity in the effects of farmers’ endowments and technology cognition on different adoption subjects and technology types. The above findings provide an in-depth explanation for activating the endogenous drivers of green agriculture development.

Well‐being and doctor visits: The mediating role of a healthy diet

AbstractSubjective well‐being is associated with many positive health outcomes. However, little is known about whether these favourable outcomes translate into a reduction in the number of doctor visits. Using large‐scale panel data from Australia, we find that individuals with high life satisfaction need fewer visits to their family doctor or general practitioner. We also present evidence on the pathways underlying this relationship, showing it is significantly explained by a person's diet. These findings remain after correcting for sample selection bias using Heckman's approach as well as controlling for various confounding factors.

The Effect of Parental Occupation and Family Arragements on Occupational Choices among Brazilian Youth

This paper estimates the added worker effect and discouraged worker effect for young people in metropolitan Brazil. Its main contribution is to reveal how the transition to unemployment by the household head impacts labor supply among youth, considering the demographic changes – especially the size and composition of family arrangements – that Brazil has undergone in recent decades. We designed multinomial probit models (with and without correction of sample selection bias) covering the main metropolitan regions of the country using data from Brazil’s Monthly Employment Survey (PME) for the years 2002 to 2015, which compiles information on monthly transitions in the Brazilian labor market. Our findings point to evidence of a greater transition of young men and women to activation in the labor market, specifically from inactivity to unemployment, in periods when the household head became unemployed. On the other hand, there is no evidence of a statistically significant discouraged worker effect impacting these individuals. Furthermore, the parents’ occupational situation seems to have little effect on the transition to activation of young males in low-income families, unlike those belonging to wealthier families.

Money talks? Obstacles to innovation in the Colombian manufacturing sector

Background and objectives: Around the world, innovation has become a central issue on the agenda of policymakers given that innovation activities are critical as a generator of competitiveness. This work adds to the literature on the elements that hinder firms’ ability to innovate. Therefore, this work analyzes the relationship between innovation propensity and innovation barriers in the Colombian manufacturing sector.
 Methods: This relationship was estimated through a Probit model using cross-sectional data provided by the National Administrative Department of Statistics (DANE). In addition, we performed the correction of the sample selection bias present in many studies on barriers to innovation as proposed in recent literature (Pellegrino and Savona, 2017). Explanatory variables were included to understand firms’ characteristics, such as firms’ size, technological intensity of the industries, among others.
 Findings: Among the relevant results is that obstacles to innovation have negative effects on innovation. More precisely, financial barriers have the greatest effect on the firms’ propensity to innovate, followed by explanatory variables, such as firms’ size, expenditures on innovation activities, and their technological intensity.
 Conclusion: We test the effect of potential innovators’ perception of the importance of diverse obstacles to innovation on their ability to produce innovative goods (or services). More precisely, we tested the assumption of non-financial and financial barriers affected firms’ innovation propensity. We found evidence of our main conjecture that financial related barriers are the most relevant and important obstacle for innovation.

Role of market participation on smallholder vegetable farmers' wellbeing: Evidence from matching approach in Eastern India

AbstractThe study identifies determinants of vegetable farmers' market participation and evaluates their impact on households' welfare. In this study, we collected 175 cross‐sectional household data and employed a probit regression model to identify the determinants of vegetable farmers' market participation. The study estimates the impact of market participation on households' income and per capita consumption using the propensity score matching approach (PSM). Subsequently, the study used the endogenous switch regression (ESR) model to control sample selection bias and unobserved heterogeneity. ESR also evaluates the robustness of PSM outcomes. Results from PSM and ESR approaches suggest that vegetable market participation positively and significantly impacts smallholders' income and consumption expenditure. The study rejects the null hypothesis that vegetable market participation does not significantly impact farmers' welfare. Results show that vegetable market participation increases the farmers' monthly income and annual per capita consumption expenditure on average by Indian rupee (INR) 8360 and INR 26,410, respectively. Findings also indicate that household size, distance to market, off‐farm income, access to training, price information, vehicle ownership, and storage systems are critical determinants of vegetable farmers' market participation. Identified determinants may support policymakers in designing suitable agricultural marketing policies to reduce market entry barriers and motivate farmers for market participation. [EconLit Citations: Q13, D60].

An Approach to Integrating a Non-Probability Sample in the Population Census

Population censuses are increasingly using administrative information and sampling as alternatives to collecting detailed data from individuals. Non-probability samples can also be an additional, relatively inexpensive data source, although they require special treatment. In this paper, we consider methods for integrating a non-representative volunteer sample into a population census survey, where the complementary probability sample is drawn from the rest of the population. We investigate two approaches to correcting non-probability sample selection bias: adjustment using propensity scores, which models participation in the voluntary sample, and doubly robust estimation, which has the property of persisting possible misspecification of the latter model. We combine the estimators of population parameters that correct the selection bias with the estimators based on a representative union of both samples. Our analysis shows that the availability of detailed auxiliary information simplifies the applied estimation procedures, which are efficient in the Lithuanian census survey. Our findings also reveal the biased nature of the non-probability sample. For instance, when estimating the proportions of professed religions, smaller religious communities exhibit a higher participation rate than other groups. The combination of estimators corrects such selection bias. Our methodology for combining the voluntary and probability samples can be applied to other sample surveys.