scholarly journals Stay Home Save Lives: A Machine Learning Approach to Causal Inference to Evaluate Impact of Social Distancing in the US

2021 ◽  
Author(s):  
Syed Muhammad Ishraque Osman ◽  
Nazmus Sakib
Keyword(s):  
Machine Learning ◽  
Policy Evaluation ◽  
Fixed Effects ◽  
Control Method ◽  
Control Group ◽  
Us States ◽  
Social Distancing ◽  
The Social ◽  
The Us ◽  
The Impact

Abstract Introduction: This study presents a machine learning based evaluation of the social distancing measures implemented in the US states. Objectives: Although there are a few studies that provide estimations of the impact of COVID-19 pandemic, there is a need for an actual policy evaluation of the already implemented social distancing measures. This paper presents an evaluation of the social distancing measures implemented by the US states. Methods: This research uses a machine learning based Generalized Synthetic Control Method. In doing so, it considers the US states that adopted early social distancing approaches as the treatment group and the states that adopted social distancing much later as the control group and it has controlled for state and time fixed effects, to cancel out the possible selection bias and endogeneity. Results: The results show that the first round of social distancing in the US is associated with lower COVID-19 infection growth rate (by -167%) when compared to the no policy intervention counterfactual. Conclusions: The findings from this policy evaluation establishes a robust scientific basis of the efficacy of social distancing measures on slowing down the contagion of a pandemic.

scholarly journals Stay Home Save Lives: A Machine Learning Approach to Causal Inference to Evaluate Impact of Social Distancing in the US

2020 ◽  
Author(s):  
Syed Muhammad Ishraque Osman ◽  
Nazmus Sakib
Keyword(s):  
Machine Learning ◽  
Fixed Effects ◽  
Control Method ◽  
Control Group ◽  
Synthetic Control ◽  
Us States ◽  
Social Distancing ◽  
Machine Learning Approach ◽  
The Us ◽  
The Impact

Abstract Although there are few studies done to provide estimations of the impact of COVID-19 pandemic, however, there is a need for an actual policy evaluation of the already implemented social distancing measures. In the US context in specific, this is especially instrumental because nearly a dozen US states are considering the reopening of the economy following anti social distancing protests. Using a machine learning based Generalized Synthetic Control Method, considering the US states that adopted early social distancing approaches as the treatment group and the states that adopted social distancing much later as the control group and controlling for state and time fixed effects (to cancel out the selection bias and endogeneity), this paper finds that social distancing is associated with lower COVID-19 infection growth rate (by 192%) when compared to the no policy intervention counterfactual.

scholarly journals Mining Google and Apple mobility data: temporal anatomy for COVID-19 social distancing

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Corentin Cot ◽  
Giacomo Cacciapaglia ◽  
Francesco Sannino
Keyword(s):  
Infection Rate ◽  
The United States ◽  
Strong Decrease ◽  
Social Distancing ◽  
Mobility Data ◽  
Actual Measure ◽  
The Us ◽  
The Impact ◽  
Political Decisions ◽  
Mobility Reduction

AbstractWe employ the Google and Apple mobility data to identify, quantify and classify different degrees of social distancing and characterise their imprint on the first wave of the COVID-19 pandemic in Europe and in the United States. We identify the period of enacted social distancing via Google and Apple data, independently from the political decisions. Our analysis allows us to classify different shades of social distancing measures for the first wave of the pandemic. We observe a strong decrease in the infection rate occurring two to five weeks after the onset of mobility reduction. A universal time scale emerges, after which social distancing shows its impact. We further provide an actual measure of the impact of social distancing for each region, showing that the effect amounts to a reduction by 20–40% in the infection rate in Europe and 30–70% in the US.

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.

scholarly journals Estimating Causal Effects When the Treatment Affects All Subjects Simultaneously: An Application

2021 ◽  
Vol 5 (2) ◽  
pp. 22
Author(s):  
Chiara Binelli
Keyword(s):  
Machine Learning ◽  
Carbon Dioxide ◽  
Global Warming ◽  
Carbon Dioxide Emissions ◽  
Control Group ◽  
Sources Of Information ◽  
Average Global Temperature ◽  
Causal Impact ◽  
The Impact ◽  
Robust Evidence

Several important questions cannot be answered with the standard toolkit of causal inference since all subjects are treated for a given period and thus there is no control group. One example of this type of questions is the impact of carbon dioxide emissions on global warming. In this paper, we address this question using a machine learning method, which allows estimating causal impacts in settings when a randomized experiment is not feasible. We discuss the conditions under which this method can identify a causal impact, and we find that carbon dioxide emissions are responsible for an increase in average global temperature of about 0.3 degrees Celsius between 1961 and 2011. We offer two main contributions. First, we provide one additional application of Machine Learning to answer causal questions of policy relevance. Second, by applying a methodology that relies on few directly testable assumptions and is easy to replicate, we provide robust evidence of the man-made nature of global warming, which could reduce incentives to turn to biased sources of information that fuels climate change skepticism.

Social Distancing Using AI and Deep Learning

2021 ◽  
Vol 9 (VII) ◽  
pp. 1816-1822
Author(s):  
Mr. Kiran Mudaraddi
Keyword(s):  
Deep Learning ◽  
Object Detection ◽  
Open Source ◽  
Suggested Method ◽  
Video Frame ◽  
Social Distancing ◽  
The Social ◽  
The Impact

The paper presents a deep learning-based methodology for detecting social distancing in order to assess the distance between people in order to mitigate the impact of the coronavirus pandemic. The input was a video frame from the camera, and the open-source object detection was pre-trained. The outcome demonstrates that the suggested method is capable of determining the social distancing measures between many participants in a video.

scholarly journals A regression discontinuity analysis of the social distancing recommendations for older adults in Sweden during COVID-19

2021 ◽  
Author(s):  
Carl Bonander ◽  
Debora Stranges ◽  
Johanna Gustavsson ◽  
Matilda Almgren ◽  
Malin Inghammar ◽  
...  
Keyword(s):  
Older Adults ◽  
Regression Discontinuity ◽  
Model Specification ◽  
Regression Discontinuity Design ◽  
Sharp Drop ◽  
Social Distancing ◽  
Disease Outcomes ◽  
The Social ◽  
National Register Data ◽  
The Impact

Objectives: To study the impact of non-mandatory, age-specific social distancing recommendations for older adults (70+ years) in Sweden on isolation behaviors and disease outcomes during the first wave of the COVID-19 pandemic. Methods: Our study relies on self-reported isolation data from COVID Symptom Study Sweden (n = 96,053) and national register data on COVID-19 hospitalizations, deaths, and confirmed cases. We use a regression discontinuity design to account for confounding factors, exploiting the fact that exposure to the recommendation was a discontinuous function of age. Results: By comparing individuals just above to those just below the age limit for the policy, our analyses revealed a sharp drop in the weekly number of visits to crowded places at the 70-year-threshold (-13%). Severe COVID-19 cases (hospitalizations or deaths) also dropped abruptly by 16% at the 70-year-threshold. Our data suggest that the age-specific recommendations prevented approximately 1,800 to 2,700 severe COVID-19 cases, depending on model specification. Conclusion: The non-mandatory, age-specific recommendations helped control the COVID-19 pandemic in Sweden.

scholarly journals Assessment of the impact of the employment roadmap on the unemployment rate of the Republic of Kazakhstan: synthetic control method

2021 ◽  
pp. 121-130
Author(s):  
T. S. Sokira ◽  
Z. T. Myshbayeva
Keyword(s):  
Unemployment Rate ◽  
Control Method ◽  
Action Plan ◽  
Control Group ◽  
Synthetic Control ◽  
Donor Pool ◽  
Synthetic Control Method ◽  
The Republic ◽  
The Impact ◽  
Weighted Combination

The purpose of the research is to assess the impact of the action plan of the Employment Roadmap on the unemployment rate in Kazakhstan.Methodology. Synthetic Control Method was used in this paper. The method, which compares one or more units exposed to the event and determines what would have happened if the unit had not been treated. In other words, this method creates a weighted combination of control states to create a single «synthetic» control group, in order to approach the counterfactual unit in Kazakhstan in the absence of a plan or Roadmap.The originality / value of the research based on the analysis, panel data from Kazakhstan and 13 donor pool countries for the period 2000-2019 were taken for modeling.Findings: As a result of the study, it was revealed that the unemployment rate would have been 2% higher in 2019 if Kazakhstan had not adopted an action plan in the form of an Employment Roadmap in 2009.

The elephant in the room of mutual forbearance

2018 ◽  
Vol 11 (2) ◽  
pp. 257-279 ◽  
Author(s):  
Burak Cem Konduk
Keyword(s):  
Airline Industry ◽  
Fixed Effects ◽  
Research Question ◽  
Fixed Effects Model ◽  
Content Type ◽  
High Performing ◽  
The Us ◽  
Mutual Forbearance ◽  
Low Performance ◽  
The Impact

PurposeThe purpose of this paper is to explain how a multi-market firm develops the motivation to forbear from competition.Design/methodology/approachA two-way fixed effects model with Driscoll and Kraay standard errors investigates the research question with panel data collected from the US scheduled passenger airline industry.FindingsThe results demonstrate that although the interaction of multi-market contact with strategic similarity impairs a firm’s forbearance from competition, the same interaction promotes it as firm performance deteriorates, supporting the hypotheses.Research limitations/implicationsPerformance explains not only how forbearance emerges out of coincidental multi-market contact but also reconciles the mixed evidence for the impact of the two-way interaction between multi-market contact and strategic similarity on forbearance.Practical implicationsAntitrust authorities should pay more attention to low performing firms than to high performing firms in their investigations. Also, managers of multi-market firms should identify multi-market rivals with low performance as targets for the initiation of forbearance.Originality/valueThis study revises the mutual forbearance theory to align it with the accumulating empirical evidence that otherwise refutes its assumption and thereby improves theory’s descriptive and predictive power.

HuDA_COVID Human Disposition Analysis during COVID-19 using Machine Learning

2022 ◽  
Vol 13 (2) ◽  
pp. 0-0
Keyword(s):  
Machine Learning ◽  
Health Status ◽  
Human Life ◽  
Age Groups ◽  
Psychological Needs ◽  
Mass Survey ◽  
Negative Effects ◽  
Methodical Approach ◽  
The Social ◽  
The Impact

Coronavirus has greatly impacted various aspects of human life, including human psychology & human disposition. In this paper, we attempted to analyze the impact of the COVID-19 pandemic on human health. We propose Human Disposition Analysis during COVID-19 using machine learning (HuDA_COVID), where factors such as age, employment, addiction, stress level are studied for human disposition analysis. A mass survey is conducted on individuals of various age groups, regions & professions, and the methodology achieved varied accuracy ranges of 87.5% to 98%. The study shows people are worried about lockdown, work & relationships. Furthermore, 23% of the respondents have not had any effect. 45% and 32% have had positive and negative effects, respectively. It is a novel study in human disposition analysis in COVID-19 where a novel weighted assignment indicating the health status is also proposed. HuDA_COVID clearly indicates a need for a methodical approach towards the human psychological needs to help the social organizations formulating holistic interventions for affected individuals.

scholarly journals The impact of a self-management educational program coordinated through WhatsApp on diabetes control

2020 ◽  
Vol 18 (2) ◽  
pp. 1841
Author(s):  
Muaed Al Omar ◽  
Sanah Hasan ◽  
Subish Palaian ◽  
Shrouq Mahameed
Keyword(s):  
Social Media ◽  
Health Literacy ◽  
Diabetes Education ◽  
Test Group ◽  
Control Group ◽  
Diabetic Patients ◽  
The Social ◽  
Media Intervention ◽  
The Impact ◽  
Hba1c Levels

Background: Social media can effectively mediate digital health interventions and thus, overcome barriers associated with face-to-face interaction. Objective: To assess the impact of patient-centered diabetes education program administered through WhatsApp on glycosylated hemoglobin (HbA1c) values, assess the correlation, if any, between health literacy and numeracy on intervention outcomes Methods: During an ‘intervention phase’ spread over six months, target diabetic patients (N=109) received structured education through WhatsApp as per the American Association of Diabetes Educators Self-Care Behaviors recommendations. The control group with an equal number of participants received ‘usual care’ provided by health professionals void of the social media intervention. Changes in HbA1c levels were recorded thrice (at baseline, 3 and 6 months) for the test group and twice (baseline and 6 months) for the control group. Change in HbA1c values were compared and statistical significance was defined at p<0.05. Baseline health literacy and diabetes numeracy were assessed for both groups (N=218) using the Literacy Assessment for Diabetes (LAD), and the Diabetes Numeracy Test (DNT), respectively, and values were correlated with HbA1c change p<0.05. Participants’ satisfaction with the intervention was also assessed. Results: The average age of respondents was 41.98 (SD 15.05) years, with a diabetes history of 10.2 (SD 8.5) years. At baseline, the average HbA1c in the control and test groups were 8.4 (SD 1.06) and 8.5 (SD 1.29), respectively. After six months, a significant drop in HbA1c value was noticed in intervention group (7.7; SD 1.35; p= 0.001); with no significance in the control group (8.4; SD 1.32; p=0.032, paired t-test). Moreover, the reduction in HbA1c was more in the test group (0.7%) than the control group (0.1%) with a difference of 0.6% which is considered clinically significant. There was no significant correlation between LAD score and HbA1c at baseline (r=-0.203, p=0.064), 3 months (r=-0.123, p=0.266) and 6 months (r=-0.106, p= 0.337) Pearson correlation. A similar result was observed with DNT, where DNT score and HbA1c at baseline, 3 months and 6 months showed no correlation (r=0.112, 0.959 and 0.886; respectively) with HbA1c levels. Eighty percent of the respondents found the social media intervention ‘beneficial’ and suggested it be used long term. Conclusions: Diabetes education via WhatsApp showed promising outcomes regardless of the level of patients’ health literacy or numeracy.

Sign in / Sign up

Export Citation Format

Share Document