Comparing Accuracy of SIR, Linear Regression, Logistic Regression, and ARIMA Models to Predict COVID-19 Case Counts on a Regional Level.
The COVID-19 pandemic has posed a significant challenge to the global community, and there have been considerable efforts to develop models to predict the behavior of the disease, particularly with regards to forecasting case counts. However, this task has been particularly challenging on a global scale, given the high variation in how different countries, and even different states, have handled the pandemic in terms of testing, mask-wearing, social distancing, and other factors. To overcome this issue, regional models have been developed that can provide more accurate predictions of case counts based on the specific conditions in a particular area. This project aimed to fit several models to COVID-19 case data on a state-by-state basis to determine which models work best in predicting case counts. The models analyzed include SIR, linear regression, logistic regression, and ARIMA time series models, with their effectiveness measured through mean squared error (MSE).