Using information on realized effects to determine prospective causal effects

The potential outcomes approach to causal inference postulates that each individual has a number of possibly latent outcomes, each of which would be observed under a different treatment. For any individual, some of these outcomes will be unobservable or counterfactual. Information about post-treatment characteristics sometimes allows statements about what would have happened if an individual or group with these characteristics had received a different treatment. These are statements about the realized effects of the treatment. Determining the likely effect of an intervention before making a decision involves inference about effects in populations defined only by characteristics observed before decisions about treatment are made. Information on realized effects can tighten bounds on these prospectively defined measures of the intervention effect. We derive formulae for the bounds and their sampling variances and illustrate these points with data from a hypothetical study of the efficacy of screening mammography.     

Response to the COVID-19: Understanding implications of government lockdown policies

The rising number of COVID-19 cases and economic implications of lockdown measures indicate the tricky balancing act policy makers face as they implement the subsequent phases of 'unlock'. We develop a model to examine how lockdown and social distancing measures have influenced the behavioral conduct of people. The current situation highlights that policy makers need to focus on bringing awareness and social restraint among people rather than going for stringent lockdown measures. We believe this work will help the policy makers gain insights into the troubled COVID-19 times ahead, and based on the estimates, they can frame policies to navigate these wild waves in the best possible way.     

Covariate selection for identifying the causal effects of stochastic interventions using causal networks

Suppose that cause-effect relationships between variables can be described by a causal network corresponding to a linear structural equation model. Kuroki and Miyakawa (2003) proposed a graphical criterion for selecting covariates to identify the causal effect of a conditional intervention. In this paper, we extend Kuroki and Miyakawa (2003) graphical criterion for selecting covariates to identify the causal effect of a stochastic intervention. Since stochastic intervention is a generalization of conditional intervention, our paper makes the results of Kuroki and Miyakawa (2003) more generally applicable.     

Non-linear autoregressive time series with multivariate Gaussian mixtures as marginal distributions

A new form of non-linear autoregressive time series is proposed to model solar radiation data, by specifying joint marginal distributions at low lags to be multivariate Gaussian mixtures. The model is also a type of multiprocess dynamic linear model, but with the advantage that the likelihood has a closed form.     

Typology of causes of poverty: The perception of Iranian farmers

Causal attributions are important mediators of future behaviour because once a cause is assigned; a commensurate action can be taken. The aims of this research were to find how Iranian farmers attribute the causes of poverty, to provide a typology of attribution of causes of poverty as perceived by Iranian farmers and to compare socio-economic characteristics and contextual conditions of farmers based on their causal attribution of poverty type. Findings revealed that 50% of respondents had structuralistic, almost 30% had individualistic, and about 20% hold fatalistic attitudes towards the causes of poverty. Farmers with individualistic attitudes towards the causes of poverty had higher quality of life, well-being, level of agricultural technology, agricultural production, used insurance more often, they had more land, income, access to agricultural extension services and practiced sustainable agriculture more often. Farmers with fatalistic attitudes towards the causes of poverty had the worst condition with regard to the above variables. However, those who had structuralistic attitudes, stood somewhere between the two previous groups. Based on findings, a number of poverty alleviation recommendations are made.     

Propensity score model specification for estimation of average treatment effects

Treatment effect estimators that utilize the propensity score as a balancing score, e.g., matching and blocking estimators are robust to misspecifications of the propensity score model when the misspecification is a balancing score. Such misspecifications arise from using the balancing property of the propensity score in the specification procedure. Here, we study misspecifications of a parametric propensity score model written as a linear predictor in a strictly monotonic function, e.g. a generalized linear model representation. Under mild assumptions we show that for misspecifications, such as not adding enough higher order terms or choosing the wrong link function, the true propensity score is a function of the misspecified model. Hence, the latter does not bring bias to the treatment effect estimator. It is also shown that a misspecification of the propensity score does not necessarily lead to less efficient estimation of the treatment effect. The results of the paper are highlighted in simulations where different misspecifications are studied.     

对纵贯数据统计分析的认识

在介绍了纵贯数据的设计思想及优缺点基础上,从统计方法的角度讨论了纵贯数据在社会科学中所能发挥的作用。纵贯数据的优点在于其可以帮助我们进行对总体异质性的识别、对因果机制干预的研究、对因果效应的研究和对状态变换的研究。以一些基于纵贯数据的研究设计为实例,文章阐述了在研究中假设与数据紧密衔接的重要性,以及利用统计方法分析纵贯数据时需要考虑的要点。但由于存在着由人类和人类行为内在变异性导致的这一无法避免的根本性缺陷,纵贯数据并不能解决所有问题。因而在纵贯数据的辅助下,研究者需要对社会现象有更深入的理解,将其进行更合理的概念化,并加以更精准的数据分析。     

北京市能源消费与经济增长关系研究

 本文利用面板协整理论和基于面板误差修正模型的Granger因果关系检验分析了北京市能源消费和经济增长的关系。面板协整检验表明北京市能源消费和经济增长之间存在长期协整关系,进一步基于面板误差修正模型的Granger因果关系检验表明北京市短期存在能源消费到经济增长的单向因果关系,长期能源消费和经济增长之间存在双向的因果关系。因此北京市在进行节能减排的工作时,必须考虑到能源消费减少对经济增长的负作用,尽可能采取提高能源利用效率的措施,包括利用财政税收优惠政策鼓励节能技术的研发,在政府采购时要求产品在生产过程中采用节能技术,更关键也是更重要的是积极探索能源价格机制改革,通过价格手段促进企业真正具备节能意识,主动节约能源,提高利用效率。     

Experimental designs in management and leadership research: Strengths,limitations, and recommendations for improving publishability

Despite the renewed interest in the use of experimental designs in the fields of leadership and management over the past few decades, these designs are still relatively underutilized. Although there are several potential reasons for this, chief among them is misunderstanding the value of these designs. The purpose of this article is to review the role of laboratory, field, and quasi-experimental designs in management and leadership research. We first discuss the primary goals of experimental studies. Next, we examine the characteristics of experimental designs and how to distinguish laboratory, field, and quasi-experiments from one another and from non-experimental studies. Following these discussions, we provide examples of each type of experimental design and discuss their relative strengths and limitations. Finally, we discuss steps that researchers can take to increase the probability of having articles reporting experiments accepted by leadership and management journals.     

Likelihood methods in randomized trials with noncompliance and subsequent nonresponse

This article considers likelihood methods for estimating the causal effect of treatment assignment for a two-armed randomized trial assuming all-or-none treatment noncompliance and allowing for subsequent nonresponse. We first derive the observed data likelihood function as a closed form expression of the parameter given the observed data where both response and compliance state are treated as variables with missing values. Then we describe an iterative procedure which maximizes the observed data likelihood function directly to compute a maximum likelihood estimator (MLE) of the causal effect of treatment assignment. Closed form expressions at each iterative step are provided. Finally we compare the MLE with an alternative estimator where the probability distribution of the compliance state is estimated independent of the response and its missingness mechanism. Our work indicates that direct maximum likelihood inference is straightforward for this problem. Extensive simulation studies are provided to examine the finite sample performance of the proposed methods.