December 20, 2020

Together, we can shape a better future using data science!

Data scientists have the best job in the world!  

COVID19 Model Forecast Comparison

Below are links to the codebase of the major global COVID19 forecasting models to compare predictions and evaluate predictive performance:

• DELPHI-MIT (Delphi) 
• Imperial College London (Imperial)
• The Institute for Health Metrics and Evaluation (IHME)
• The Los Alamos National Laboratory (LANL)
• SIKJalpha
• Youyang Gu (YYG)

Considering the poor predictive performance of all COVID19 models, there is lively debate about the usefulness of forecasts and alternative scenarios for policy decision makers when dealing with complex pathogens and potential policy responses in high causal density environments.

Many argue that doomsday predictions scared both political leaders and the public and caused bad or suboptimal policy responses that caused significant damage to society.

Should the creators and institutions of such doomsday predictive models have exposure to civil and criminal liability?

New Data Science News Articles: 

• Opening the ‘Black Box’ of Artificial Intelligence
• Ten Commandments of Fake Science
• Radical Empiricism and Machine Learning Research
• Price of Panic - Lockdowns Cost Lives

New Books from DSA Store:

• Probabilistic Graphical Models: Principles and Techniques
• Building Machine Learning Systems with Python

New DSA Videos:

• Statistical Inference 03
• Statistical Inference 04

New DSA Resources:

• SMOTEBoost: Improving Prediction of the Minority Class in Boosting
• Journey into Bayesian Networks

© 2020 Data Science Association, Inc. — All Rights Reserved.