Using Kalman Filter to Predict COVID19 Spread

This work presents the implementation of an online real-time Kalman filter algorithm to predict the spread of COVID19 per given region. Coronavirus(COVID19 or SARS-CoV-2) has recently caused major worldwide concern. As the number of coronavirus cases reportedly increases, the spread of COVID19 is a serious threat to global health. In this work, we will try to predict the spread of coronavirus for each one of the infected regions. Fitting time series analysis and statistical algorithms to produce the best short term and long term prediction. An adaptive online Kalman filter provides us very good one-day predictions for each region.

Links to Medium article can be found here

• Using Kalman Filter to Predict Corona Virus Spread (Feb 22)
• Coronavirus Spread Updated Prediction Using Kalman Filter (Mar 1)

Kalman Filter을 이용한 대한민국의 코로나19 확진자/사망자/완치자 예측

대한민국의 경우에 한정해서 Ran Kremer의 모델로 확진자 수 / 사망자 수 / 완치자 수의 경향을 예측해보려고 합니다.

• (클릭) Google Colab으로 예측 모델을 직접 실행시켜보세요
• (클릭) 국문으로 번역된 해설을 확인해보세요