Generalized Synthetic Control Method for Causal Inference with Time-Series Cross-Sectional Data

2015 ◽  
Author(s):  
Yiqing Xu
Keyword(s):  
Time Series ◽  
Causal Inference ◽  
Control Method ◽  
Synthetic Control ◽  
Cross Sectional ◽  
Synthetic Control Method

scholarly journals Generalized Synthetic Control Method: Causal Inference with Interactive Fixed Effects Models

2017 ◽  
Vol 25 (1) ◽  
pp. 57-76 ◽  
Author(s):  
Yiqing Xu
Keyword(s):  
Causal Inference ◽  
Fixed Effects ◽  
Control Method ◽  
The United States ◽  
Control Group ◽  
Synthetic Control ◽  
Synthetic Control Method ◽  
Varying Coefficients ◽  
Fixed Effects Models ◽  
Interactive Fixed Effects

Difference-in-differences (DID) is commonly used for causal inference in time-series cross-sectional data. It requires the assumption that the average outcomes of treated and control units would have followed parallel paths in the absence of treatment. In this paper, we propose a method that not only relaxes this often-violated assumption, but also unifies the synthetic control method (Abadie, Diamond, and Hainmueller 2010) with linear fixed effects models under a simple framework, of which DID is a special case. It imputes counterfactuals for each treated unit using control group information based on a linear interactive fixed effects model that incorporates unit-specific intercepts interacted with time-varying coefficients. This method has several advantages. First, it allows the treatment to be correlated with unobserved unit and time heterogeneities under reasonable modeling assumptions. Second, it generalizes the synthetic control method to the case of multiple treated units and variable treatment periods, and improves efficiency and interpretability. Third, with a built-in cross-validation procedure, it avoids specification searches and thus is easy to implement. An empirical example of Election Day Registration and voter turnout in the United States is provided.

scholarly journals Synthetic control methodology as a tool for evaluating population-level health interventions

2018 ◽  
Vol 72 (8) ◽  
pp. 673-678 ◽  
Author(s):  
Janet Bouttell ◽  
Peter Craig ◽  
James Lewsey ◽  
Mark Robinson ◽  
Frank Popham
Keyword(s):  
Public Health ◽  
Time Series ◽  
Time Series Analysis ◽  
Observational Data ◽  
Control Method ◽  
Health Interventions ◽  
Synthetic Control ◽  
Public Health Interventions ◽  
Synthetic Control Method ◽  
Control Units

BackgroundMany public health interventions cannot be evaluated using randomised controlled trials so they rely on the assessment of observational data. Techniques for evaluating public health interventions using observational data include interrupted time series analysis, panel data regression-based approaches, regression discontinuity and instrumental variable approaches. The inclusion of a counterfactual improves causal inference for approaches based on time series analysis, but the selection of a suitable counterfactual or control area can be problematic. The synthetic control method builds a counterfactual using a weighted combination of potential control units.MethodsWe explain the synthetic control method, summarise its use in health research to date, set out its advantages, assumptions and limitations and describe its implementation through a case study of life expectancy following German reunification.ResultsAdvantages of the synthetic control method are that it offers an approach suitable when there is a small number of treated units and control units and it does not rely on parallel preimplementation trends like difference in difference methods. The credibility of the result relies on achieving a good preimplementation fit for the outcome of interest between treated unit and synthetic control. If a good preimplementation fit is established over an extended period of time, a discrepancy in the outcome variable following the intervention can be interpreted as an intervention effect. It is critical that the synthetic control is built from a pool of potential controls that are similar to the treated unit. There is currently no consensus on what constitutes a ‘good fit’ or how to judge similarity. Traditional statistical inference is not appropriate with this approach, although alternatives are available. From our review, we noted that the synthetic control method has been underused in public health.ConclusionsSynthetic control methods are a valuable addition to the range of approaches for evaluating public health interventions when randomisation is impractical. They deserve to be more widely applied, ideally in combination with other methods so that the dependence of findings on particular assumptions can be assessed.

scholarly journals Essays in policy evaluation and labor economics

2021 ◽  
Author(s):  
◽  
Zhiyang You
Keyword(s):  
Control Method ◽  
Control Group ◽  
Synthetic Control ◽  
Cross Sectional ◽  
Synthetic Control Method ◽  
Assimilation Effect ◽  
The Law ◽  
The Difference ◽  
Cohort Quality ◽  
The U.S

This dissertation contains three chapters. The first chapter evaluates the effect of a gun control act in California. State legislators in the U.S. are striving to curb gun violence. A common approach is to extend the existing firearms ban list. This paper examines the effect of legislation restricting sales of selected firearms in California using the synthetic control method. This case study method forms a synthetic unit using a linear combination of other states in the U.S. as the control group. The results show substantial increases in firearm sales in California from the point of passage until the law becomes effective. After the surge ends when the law becomes effective, the sale of firearms is only moderately affected thereafter. This paper also creates robustness checks to confirm that the synthetic control method is working properly with low firearm density in California, which calls into question some of the assumptions underlying the synthetic control method. The Difference-in-Difference regression reaches the same conclusion. The second chapter focuses on immigrant assimilation in the U.S. Assimilation is the process in which immigrants improve earnings as they become more adapted to the host country society. Cross-sectional studies show that immigrants have lower earnings upon arrival and faster earnings growth compared to natives. Longitudinal studies conclude that estimates based on cross-sectional data are positively biased due to decreasing cohort quality and negatively selected outmigration. I reproduce such estimates with recent U.S. data. The estimates would appear to show "bias," as inclusion of cohort fixed effects alter estimates. However, in contrast to expectations based on the current literature, decreasing cohort quality and outmigration do not explain the difference. Next, I apply a non-parametric method to make the wage distributions visually comparable across cohorts and time. I find that the linear specification of assimilation is misleading. Finally, I revisit the classic model with a quadratic assimilation term and expand it to explore the assimilation process's heterogeneity. I find that the "bias" disappears with a quadratic assimilation effect. The assimilation effect is sensitive to age at arrival and country of origin. The third chapter considers an unexplained puzzle in one of the most widely used public datasets in the U.S. The American Community Survey (ACS) replaced the Decennial Census as the primary data source for identifying immigrants' socioeconomic characteristics. This paper focuses on cohort analysis, in which a cohort combines immigrants arriving in a given year from surveys in multiple years. Tracking the sizes of cohorts from 2006 to 2019 using the ACS, we observe an abnormal increase in cohort size in the 10th and 20th years since arrival. Two hypotheses are tested, population estimate structural break and the renewal of green card. Neither appears to explain the puzzle.

Applying Synthetic Control Method to Estimate the Impact of the Lava Jato Operation on the Worldwide Governance Indicators in Brazil

2021 ◽  
pp. 2631309X2110178
Author(s):  
Eduardo Carvalho Nepomuceno Alencar ◽  
Bryant Jackson-Green
Keyword(s):  
Rule Of Law ◽  
Control Method ◽  
Synthetic Control ◽  
Synthetic Control Method ◽  
The Rule Of Law ◽  
Governance Indicators ◽  
Regulatory Quality ◽  
Legal Judgments ◽  
The Impact ◽  
Control Of Corruption

In 2014, the most prominent anti-corruption investigation in Latin America called Lava Jato, exposed a Brazilian corruption scheme with reverberations in 61 countries, resulting in legal judgments for nearly 5 billion USD in reimbursements thus far. This article applies the synthetic control method on data from 135 countries (2002–2018) to test the hypothesis that Lava Jato impacts the Worldwide Governance Indicators in Brazil. The findings reveal that Lava Jato negatively affects control of corruption, the rule of law, and regulatory quality. There are signs of possible improvement in at least the corruption and the rule of law measures. This paper brings value to the criminological body of literature, notably lacking in the Global South.

scholarly journals When is a Match Sufficient? A Score-based Balance Metric for the Synthetic Control Method

2020 ◽  
Vol 8 (1) ◽  
pp. 209-228
Author(s):  
Layla Parast ◽  
Priscillia Hunt ◽  
Beth Ann Griffin ◽  
David Powell
Keyword(s):  
Los Angeles ◽  
Control Method ◽  
Control Unit ◽  
Treated Group ◽  
Control Group ◽  
Synthetic Control ◽  
Donor Pool ◽  
Synthetic Control Method ◽  
The Mean ◽  
The Impact

AbstractIn some applications, researchers using the synthetic control method (SCM) to evaluate the effect of a policy may struggle to determine whether they have identified a “good match” between the control group and treated group. In this paper, we demonstrate the utility of the mean and maximum Absolute Standardized Mean Difference (ASMD) as a test of balance between a synthetic control unit and treated unit, and provide guidance on what constitutes a poor fit when using a synthetic control. We explore and compare other potential metrics using a simulation study. We provide an application of our proposed balance metric to the 2013 Los Angeles (LA) Firearm Study [9]. Using Uniform Crime Report data, we apply the SCM to obtain a counterfactual for the LA firearm-related crime rate based on a weighted combination of control units in a donor pool of cities. We use this counterfactual to estimate the effect of the LA Firearm Study intervention and explore the impact of changing the donor pool and pre-intervention duration period on resulting matches and estimated effects. We demonstrate how decision-making about the quality of a synthetic control can be improved by using ASMD. The mean and max ASMD clearly differentiate between poor matches and good matches. Researchers need better guidance on what is a meaningful imbalance between synthetic control and treated groups. In addition to the use of gap plots, the proposed balance metric can provide an objective way of determining fit.

Sign in / Sign up

Export Citation Format

Share Document