Data Analysis and Visualization

Beating COVID-19: Policy Stringency Trajectory Peers

By Parvathy Krishnan, and Kai Kaiser

The threat of COVID-19 contagion has led countries across the world to implement countervailing policy actions. Policy measures across the globe have implied significant restrictions for ground, air, sea mobility, as well as commerce and recreation activities.

A “trinity” of measures — COVID-19 cases, government policy actions, and changes in mobility — emerged as key headline metrics for how the pandemic has evolved across countries. But as the initial focus was on ramping up emergency response, a growing number of countries are now entering hopefully phases of recovery.

The country visualizations by focused on the evolution of newly reported cases over time. This visualization groups country by whether they are winning, nearly there, or needing action by this measure. Many countries, particularly in East Asia and Europe, are all now in the process of re-opening their economies after imposing nation-wide lockdown.

The data potentially also allow us to identify policy stringency peers: countries that currently seem to have a similar evolution of policy stringency trajectories. The Oxford COVID-19 Government Response Tracker (OxCGRT) systematically collects information on several different common policy responses that governments have taken to respond to the pandemic on 17 indicators including but not limited to travel restrictions.

The stringency index is reported in the range of 0 to 100 with the value of 100 indicating a nation-wide total lockdown. Different regions around the globe have a different range of Stringency Index, with countries in East Asia and the Pacific having a higher mean compared to the other regions. The OxGGRT tries to ensure that all countries are updated at least once a week, and most are updated more frequently.

Range of Policy Stringency across the regions (Data as on 10th June 2020)

Policy stringency does not necessarily mean reductions in mobility. Voluntary behavior across countries appears to be a significant part of the story. But reductions in policy stringency seem to suggest that governments feel they have tackled at least the first wave of COVID-19.

Beating COVID-19: Vietnam

Vietnam, one of the countries in the East Asia and Pacific region, successful in its fight against the virus, is reducing its policy stringency (particularly at the domestic level), seeing a return to near-fully open mobility within the country. This is evident in the visualization below showing the stringency index and google mobility changes over time in Vietnam.

Vietnam Policy Stringency Index

Comparing Policy Stringency and Google Mobility Data shows a substantial increase in the number of people staying in residential locations whereas all other sectors show a decline during the lockdown period, and now going back to pre-COVID baselines.

Policy Stringency Index and percentage change in baseline of Google Mobility Data (June 12, 2020)

Finding Policy Stringency Peers

To see if countries like Vietnam have policy peers by the shape of their policy stringency trajectory, an algorithm was developed to show peers in terms of their trajectories. The peer algorithm works by clustering the stringency data per country to identify countries with a similar shape of the curve. The approach is documented on the public repository Stringency Index Peer Analysis.

The figure below suggests the peers based on this matching for Vietnam as of May 25, 2020. The peers are identified based on the shape of the curve rather than at the value of the index. This is done to identify countries that are relaxing or tightening lockdown synchronously rather than identifying countries that are at the same value of stringency index. Alternatives to this index could give more weight to the recent value of stringency index, and also by including the level along with the shape of the curve. The COVID-19 Observatory link also allows the peer finder to be identified for any country of interest covered by Oxford.

Stringency Peers of Vietnam

Finding Friends

If reductions in policy stringency reflect a sense on the part of different national authorities that they have beat the first wave of COVID-19, does this provide some insights into how to stimulate both national and international recovery? Re-establishing flight links between countries that are at different stages of the COVID-19 contagion response is challenging. But promoting safe recovery among peers in the COVID-19 policy response cycle can be more realistic.

The highest gains from re-establishing peer links among countries who are relaxing stringency policies are among countries where for example tourism and business travel before COVID-19 were significant. To address this part of the prioritization puzzle, we next analyze international and domestic flight patterns data, as illustrated for Vietnam. This suggests corridors that may be especially ripe for opening.

Stay tuned for this analysis on the flight patterns data.

The matching algorithm can be found on public repository Stringency Index Peer Analysis. The team welcomes any feedback and thoughts to improve this work, which is based fully on open data. The findings expressed in this blog are those of the authors and do not purport to reflect the opinions or views of any of the author's organizations.




Lead Data Scientist | CTO at Analytics for a Better World | Public Sector Consultant

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

Predictiong missing values with Linear Regression

Daily Newsletter — 21st December 2020

Why MSBI Is The Ultimate Tool

21 CSCW 2021 papers to read

COVID-19 Statistical Risk Classification

Using Bagging and Boosting to Improve Classification Tree Accuracy

Logistic Regression with PyTorch

Two different doors (options). One is yellow and one is blue.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Parvathy Krishnan

Parvathy Krishnan

Lead Data Scientist | CTO at Analytics for a Better World | Public Sector Consultant

More from Medium

Supplier Selection with TOPSIS Method (Python,Qlik Sense)

Society 5.0 and the role of Data

Exploratory Analysis on INST class prerequisites

Demystifying Data Unions