Unobserved Factors Research Articles

Empowering Greenhouse Cultivation: Dynamic Factors and Machine Learning Unite for Advanced Microclimate Prediction

Climate change has led to more frequent extreme weather events such as heatwaves, droughts, and storms, which significantly impact agriculture, causing crop damage. Greenhouse cultivation not only provides a manageable environment that protects crops from external weather conditions and pests but also requires precise microclimate control. However, greenhouse microclimates are complex since various heat transfer mechanisms would be difficult to model properly. This study proposes an innovative hybrid model (DF-RF-ANN), which seamlessly fuses three components: the dynamic factor (DF) model to extract unobserved factors, the random forest (RF) to identify key input factors, and a backpropagation neural network (BPNN) to predict greenhouse microclimate, including internal temperature, relative humidity, photosynthetically active radiation, and carbon dioxide. The proposed model utilized gridded meteorological big data and was applied to a greenhouse in Taichung, Taiwan. Two comparative models were configured using the BPNN and the Long short-term memory neural network (LSTM). The results demonstrate that DF-RF-ANN effectively captures the trends of the observations and generates predictions much closer to the observations compared to LSTM and BPNN. The proposed DF-RF-ANN model hits a milestone in multi-horizon and multi-factor microclimate predictions and offers a cost-effective and easily accessible approach. This approach could be particularly beneficial for small-scale farmers to make the best use of resources under extreme climatic events for contributing to sustainable development goals (SDGs) and the transition towards a green economy.

Air quality and employee performance in teams: Evidence from the NFL

Emerging research documents the impact of poor air quality on employee performance in a number of settings, including sport. Unlike other settings, sports teams have limited ability to influence previously scheduled games, making air quality exogenous to unobservable factors affecting game outcomes. We link play-level data from National Football League (NFL) games to data from air quality monitoring stations located near NFL stadiums. Results using data from about 90,000 offensive plays in 900 NFL games indicate that yards gained per offensive play increase with PM2.5 concentration. Defensive players are relatively more affected by air quality than offensive players.

Technological innovation and the quality of environmental information disclosure: Evidence from a-share companies listed on China's Shanghai and Shenzhen stock exchanges

High costs of technological innovation and insufficient technological innovation capabilities may impede the ability of firms to communicate relevant environmental protection information to the public. This study aims to analyze the impact of technological innovation on the quality of environmental information disclosure conducted by firms. A non-balanced panel dataset comprising A-share listed companies in Shanghai and Shenzhen from 2007 to 2020 was subjected to ordinary least squares (OLS) regression analysis in the study. The model included year-fixed and industry-fixed effects to eliminate the influence of unobservable factors during the sample period. The results revealed a statistically significant positive relationship between the quality of environmental information disclosure by enterprises and the quantity and quality of technological innovation. The study's findings indicate that technological innovation offers a more sustainable path for the advancement of environmental information disclosure. By utilizing environmentally friendly and energy-saving technologies, the negative impact of the environment on technological innovation can be mitigated, leading to a more sustainable approach to environmental information disclosure. The practical implications of this study include fostering innovation to enhance disclosure quality, which benefits policymaking and industry practices. The understanding of the positive relationship between technological innovation and environmental information disclosure is enriched by the empirical evidence from this study, emphasizing its role in promoting sustainability and transparency.

Proximal Reinforcement Learning: Efficient Off-Policy Evaluation in Partially Observed Markov Decision Processes

In applications of offline reinforcement learning to observational data, such as in healthcare or education, a general concern is that observed actions might be affected by unobserved factors, inducing confounding and biasing estimates derived assuming a perfect Markov decision process (MDP) model. In “Proximal Reinforcement Learning: Efficient Off-Policy Evaluation in Partially Observed Markov Decision Processes,” A. Bennett and N. Kallus tackle this by considering off-policy evaluation in a partially observed MDP (POMDP). Specifically, they consider estimating the value of a given target policy in an unknown POMDP, given observations of trajectories generated by a different and unknown policy, which may depend on the unobserved states. They consider both when the target policy value can be identified the observed data and, given identification, how best to estimate it. Both these problems are addressed by extending the framework of proximal causal inference to POMDP settings, using sequences of so-called bridge functions. This results in a novel framework for off-policy evaluation in POMDPs that they term proximal reinforcement learning, which they validate in various empirical settings.

Off-Farm Employment, Outsourced Machinery Services, and Farmers’ Ratoon Rice Production Behavior: Evidence from Rice Farmers in Central China

Promoting ratoon rice is a critical measure for ensuring food security in China. Understanding the mechanism underlying farmers’ decision-making involving ratoon rice production may contribute to the design and implementation of extension policies. This study examined the impact of off-farm employment and outsourced machinery services on farmers’ ratoon rice production behavior. We used a representative household survey of 1752 rice farmers in Hubei province in central China and a multinomial endogenous treatment effect model to address potential self-selection biases from both observable and unobservable factors. Our estimates suggest that the probability of ratoon rice production decreases by 17.2% for farmers with off-farm employment, while the probability of ratoon rice production increases by 78.9% and 57% for farmers with outsourced machinery services and those with simultaneous off-farm employment and outsourced machinery services, respectively. Moreover, we found that outsourced machinery services can mitigate the negative impact of agricultural labor aging and feminization on ratoon rice production to some extent. Outsourced machinery services have regional heterogeneity effects as farmers in non-plain areas are more likely to engage in ratoon rice production.

The effect of intergenerational mobility on family education investment: evidence from China

The discussion of inequality has been an enduring topic in sociology and economics. With the intensification of inequality, an increasing amount of research has begun to focus on the impact of inequality on various aspects of the economy and society. However, research on how inequality affects families’ education investment behavior currently remains relatively scarce. This study contributes to filling this gap by presenting one of the first analyses of the effect of intergenerational mobility-based opportunity inequality on family education investment. Specifically, based on a Chinese population sample survey conducted in 2015 and the China Family Panel Studies survey conducted in 2018, this paper measures the intergenerational mobility of regions using an index of intergenerational educational rank correlation, and it uses extracurricular tutoring expenses to measure families’ investment in their children’s education. The benchmark regression results show that intergenerational mobility significantly negatively impacts family education investment, with the average family education investment decreasing by 25.75 percent for every 0.1-unit increase in intergenerational mobility. This negative effect remains significant after robustness tests, such as replacing the explanatory variables and dependent variables, considering the influence of important omitted variables, evaluating the impact of unobservable factors, and introducing an instrumental variable for two-stage least squares regression analysis. In addition, this negative impact is more prevalent among families with high socioeconomic status, while it is not significant in families with low socioeconomic status. The reason is that families with low socioeconomic status face greater credit constraints and intergenerational mobility incentives. Furthermore, an examination of the mechanisms involved reveals that although the improvement in intergenerational mobility may increase people’s confidence in investing, it ultimately reduces family education investment by lowering excessive anxiety and the extent of status-seeking behavior among families. According to the analysis, promoting equality of opportunity could mitigate China’s negative educational competition and facilitate the realization of the “Double Reduction” policy.

The Evolution of the Returns to Schooling for Refugees and Other Immigrants in the United States

Previous research has shown that refugees have lower earnings than other immigrants with comparable human capital. The differences in the degree of correlation between schooling and unobservable factors across the two groups may contribute to this gap. This paper investigates the returns to schooling for refugee and non-refugee men in the United States. The results show that the returns to schooling are lower for refugees upon arrival. This gap persists over two decades in the country and is robust to specifications that conduct sensitivity analysis, suggesting that differences in unobservable characteristics that are important in immigrants’ labor market success, particularly adaptability traits, drive the initial gap in the returns to schooling and how the gap changes over time.

Order estimation for a fractional Brownian motion model of glucose control

When a subject is at rest and meals have not been eaten for a relatively long time (e.g. during the night), presumably near-constant, zero-order glucose production occurs in the liver. Glucose elimination from the bloodstream may be proportional to glycemia, with an apparently first-order, linear elimination rate. Besides glycemia itself, unobserved factors (insulinemia, other hormones) may exert second and higher order effects. Random events (sleep pattern variations, hormonal cycles) may also affect glycemia. The time-course of transcutaneously, continuously measured glycemia (CGM) thus reflects the superposition of different orders of control, together with random system error. The problem may be formalized as a fractional random walk, or fractional Brownian motion. In the present work, the order of this fractional stochastic process is estimated on night-time CGM data from one subject.

Couple and individual willingness to take risks

This study analyses the relationship between the willingness to take risk (WTR) of a sentimental couple and its individual components. Using a survey-based measure collected in a lab experiment with 126 couples, we estimate a joint model for explaining female, male, and couple WTR. We control for socio-demographic characteristics and personality traits in the individual risk specifications and for the length of the relationship in the joint risk specification. We find that individual WTR is related to personality more than to socio-demographic variables. Couple WTR is equally determined by the individual WTR of each partner, once endogeneity arising from unobservable common factors is considered. This implies that risk-averse (risk-tolerant) individuals appear to be willing to take more (less) risk when behaving with the partner than he/she would like when behaving individually.

Class size and teacher effects on non-cognitive outcomes in grades K-3: a fixed effects analysis of ECLS-K:2011 data

This study examines the association between class size, teacher characteristics and five non-cognitive student outcomes (i.e., self-control, interpersonal skills, approaches to learning, externalizing and internalizing problem behaviors) in grades K-3. Individual fixed-effects, that control for observed and unobserved time-invariant factors, including student and school time-constant variables, are employed to analyze national data from the Early Childhood Longitudinal Study, Kindergarten Class of 2010–11 (ECLS-K:2011). Results suggest a significant linear association between reducing class size and interpersonal skills. Teacher experience is saliently, positively and significantly related with student self-control, interpersonal skills, and approaches to learning, and negatively associated with externalizing problem behaviors. Teacher education and certification were not associated with any of the five non-cognitive outcomes. Changing schools had a negative effect on student’s self-control.

COVID-19 and visitation to Central Park, New York City.

Central Park is an iconic feature of New York City, which was the first and one of the hardest hit cities in the United States by the Coronavirus. State-level stay-at-home order, raising COVID-19 cases, as well as the public's personal concerns regarding exposure to the virus, led to a significant reduction of Central Park visitation. We utilized extensive cellphone tracking data to conduct one of the pioneering empirical studies assessing the economic impact of COVID-19 on urban parks. We integrated the difference-in-difference (DID) design with the recreation-demand model. The DID design aids in identifying the causal impacts, controlling for unobservable factors that might confound the treatment effects of interest. Concurrently, the recreational demand model examines the driving factors of visitation changes and enables us to estimate the welfare changes experienced by New York City's residents. Our findings shine a light on the substantial, yet often overlooked, welfare loss triggered by the pandemic. The analysis indicates that the pandemic resulted in a 94% reduction in visitation, corresponding to an annual consumer surplus loss of $450 million. We noted a rebound in visitation following the initial outbreak, influenced by shifts in government policy, weather conditions, holiday periods, and personal characteristics.

Investigating injury severity interaction between urban and urban-rural fringe areas: a grouped random parameters seemingly unrelated bivariate probit approach

Urban-rural integration is the critical process of urbanization, while during this process there are various corresponding problems generated, safety is the most important one, especially at urban-rural fringe areas due to mixed traffic flow and complicated land use. This study intended to investigate the injury severity interaction between urban area and urban-rural fringe area, and explore the correlation within injury severity levels and heterogeneity attributed to unobserved factors. To address the correlation and heterogeneity issues, a grouped random parameters seemingly unrelated bivariate (SUB) probit model was proposed, in which the SUB probit model addressed the correlation of residuals, while the random parameters model accommodated the heterogeneity due to unobserved factors. By comparing the pooled with random parameters models, the results showed that random parameters SUB probit model performed better than the pooled one, and the dataset collected from 2013 to 2017 in Chengdu, China was adopted to illustrate the proposed model. It is found that crash location, speed limit and age of person injured are significant for injury severity in urban and urban-rural fringe areas, but crash form plays a significant role in urban area while number of persons involved should be paid more attention due to injury severity in urban-rural fringe area. Some empirical suggestions are presented to improve the safety in urban and urban-rural fringe areas.

INTERACTIVE EFFECTS PANEL DATA MODELS WITH GENERAL FACTORS AND REGRESSORS

This paper considers a model with general regressors and unobservable common factors. An estimator based on iterated principal component analysis is proposed, which is shown to be not only asymptotically normal, but under certain conditions also free of the otherwise so common asymptotic incidental parameters bias. Interestingly, the conditions required to achieve unbiasedness become weaker the stronger the trends in the factors, and if the trending is strong enough, unbiasedness comes at no cost at all. The approach does not require any knowledge of how many factors there are, or whether they are deterministic or stochastic. The order of integration of the factors is also treated as unknown, as is the order of integration of the regressors, which means that there is no need to pre-test for unit roots, or to decide on which deterministic terms to include in the model.

Urban Economic Efficiency, Environmental Factors, and Digital Finance: Impacts on Sustainable Development in Chinese Cities

This study investigates the potential of digital finance to mitigate the environmental impact of economic development. Utilizing panel data from 290 Chinese cities between 2011 and 2019, we employ a two-way fixed effects model to control for unobservable city characteristics and time-invariant macro factors. Our findings reveal that cities with higher proportions of value-added tax and secondary industries (mainly for manufacturing) tend to experience increased air pollution. Conversely, cities characterized by higher economic efficiency exhibit better air quality, underscoring the significance of prioritizing technology-intensive sectors during economic development. Pollution control technologies, such as centralized wastewater treatment systems and harmless treatment systems for household waste, also play positive roles. Moreover, technological and educational investments correlate with increased urban green coverage and lower air pollution, highlighting the pivotal role of technology development and residents’ literacy in fostering sustainable practices. Although we initially observed a positive association between the index of digital financial technology and air pollution, which could be attributed to the influence of economic growth, our study employs instrumental variable regression to unveil their true correlation. The results suggest the application of digital financial technology may enhance environmental awareness, contributing to the mitigation of air pollution.

Social safety nets and productivity outcomes: Evidence and implications for Bangladeshi rice growers

AbstractUsing national representative cross‐sectional data from 8846 rice growing plots in Bangladesh, we analyze the impact of social safety net programs (SSNPs) on rice farmers' productivity and efficiency. This study applied the propensity score matching (PSM) approach to correct the possible self‐selection biases arising from observed and unobserved factors affecting a household's SSNP participation decisions. The stochastic frontier production function is estimated separately for participants and non‐participants to compare yield and technical efficiency. Our results show that the output elasticities of all inputs except land and fertilizer are significantly higher for non‐participants than participants. The SSNP participants are estimated to have 1.95% higher technical efficiency than the non‐participants, while their frontier yield is 11.75% higher. The average treatment effect on treated (ATET) also confirms the yield and efficiency‐enhancing role of SSNPs, particularly in climate‐stressed areas. The study argues for widening the SSNP coverage to increase the marginal farmers' efficiency in climate‐vulnerable areas. Other policy interventions include policies to increase farmers' access to land through the proper functioning land market, education, providing literacy training, and different production inputs.

“Transformation or not?” the effect of broadcasters' role orientation transformation on their direct income

PurposeIn the actual livestreaming background, to obtain more income, some broadcasters will transform their original single role orientation into mixed one. This research study aims to conduct an empirical study on the influence of the broadcasters' role orientation transformation on the viewers' tipping behavior.Design/methodology/approachThe authors collect data from Kuai, a leading online live streaming service provider in China. The dataset includes 175,701 live streaming data from 971 broadcasters in 7 months. To avoid unobservable factors, the authors adopt two difference-in-differences (DID) models to estimate the effect of two kinds of broadcaster's role orientation transformation on the broadcaster’s direct income separately. And the authors use the Heckman-type correction to solve broadcasters’ self-selected problem.FindingsThe authors evaluated that there is a U-shape relationship between the broadcasters' role orientation transformation and their direct income. The broadcasters' direct income experienced a sharp decline for a short period of time after transformation and followed by a rise after a period of adaptation. And for broadcasters with different genders and amounts of fans, the influence degree of role orientation transformation is various.Originality/valueThis paper provides a fresh usage of the regulatory engagement theory in the brand new information communication technology. And it also explores the boundary effect of the participating object's self-factors in the regulatory engagement theory. Besides, this paper expands the research of livestreaming into natural background. Such results also provide operable suggestions for the livestream platform, the broadcaster himself and the enterprises who want to employ some broadcasters to recommend their products.

How sensitive are matching estimates of active labor market policy effects to typically unobserved confounders?

Using a rich and unique combined administrative-survey dataset, this paper explores how sensitive propensity score (PS) matching estimates of Active Labor Market Policies (ALMPs) based on the selection-on-observables assumption are to typically unobserved covariates. Using a sample of German unemployed welfare recipients, the analysis shows that typically unobserved factors such job search behavior, concessions willing to make for a job as well as (mental) health are in fact relevant confounders. However, results also show that matching on the PS using only typically observed covariates reduces imbalance in terms of typically unobserved covariates by about 46 percent in this setting. In line with this finding, the inclusion of typically unobserved covariates yields very similar estimates to estimates based on a standard specification. Hence, a standard matching approach based on rich and high quality administrative data appears to be sufficient to obtain estimates that are rather robust to unobserved confounding.

Adolescent network positions and memory performance in adulthood: Evidence from sibling fixed effects models with sociometric network data

Although research has explored social factors influencing memory performance during adolescence, the impact of adolescent social network positions remains largely unknown. This study examines whether adolescent network position is associated with memory performance in adulthood, while also considering potential gender differences. The study used a sibling sample from the National Longitudinal Study of Adolescent to Adult Health (N = 2462) and employed sibling fixed effects models to account for unobserved family background factors, such as genetics, parental characteristics, family environment, and childhood neighborhood. Four dimensions of adolescent network position—i.e., popularity, sociality, degree centrality, and closeness centrality—were sociometrically assessed in schools. Memory performance in adulthood was measured using the Rey Auditory Verbal Learning Test. The sibling fixed effects estimates indicate that sociality, degree centrality, and closeness centrality are significantly associated with increased memory performance in adulthood, even after controlling for unobserved family heterogeneity as well as a set of individual-level covariates. In contrast, controlling for unobserved family heterogeneity attenuated the association for popularity, making it statistically insignificant. This study also provides evidence of gender differences in the association between social network position and memory performance. The associations for popularity, sociality, and degree centrality are more pronounced among men than women. This study’s findings highlight the importance of adolescent network positions as social determinants in shaping cognitive outcomes over the life course. Interventions that encourage positive peer interactions and reduce social isolation during adolescence may help improve cognitive health in the population.

Are students time constrained? Course load, GPA, and failing

Given the simultaneous rise in time-to-graduation and college GPA, it may be that students reduce their course load to improve their performance. Yet, evidence to date only shows increased course loads increase GPA. We provide a model showing many unobservable factors – beyond student ability – can generate a positive relationship between course load and GPA unless researchers control student schedules. West Point regularly implements the ideal experiment by randomly modifying student schedules with additional training courses. Using 19 years of administrative data, we provide the first causal evidence that taking more courses reduces GPA and increases course failure rates, sometimes substantially.

Climate regulation costs and firms’ distress risk

AbstractIn 2013, the European Union's Emission Trading Scheme (EU‐ETS) entered Phase III. The majority of emission permits in Phase III are auctioned instead of being allocated for free as in Phases I and II. Using a difference‐in‐differences method, we show that this change has led to an increase in the financial distress risk of the EU‐ETS‐regulated firms when compared to unregulated firms, suggesting that the EU‐ETS imposes a significant financial burden on regulated firms. This result is robust to an array of validation tests, alleviating concerns that it is driven by unobserved factors. In additional analyses we show that the increase in distress risk of regulated firms during Phase III can be explained by, (i) an additional climate regulation cost to purchase pollution permits and (ii) a low average environmental score that possibly (via high sustainability risk) lowers investors expectations regarding firms’ performance. Our findings also show that the distress risk increase is higher for regulated firms operating within countries with lower control of corruption, government effectiveness, political stability, regulatory quality, rule of law, and voice accountability before the EU‐ETS implementation.