2021.0027

This archive is distributed in association with the INFORMS Journal on Computing under the [MIT License (LICENSE).

The software and data in this repository are a snapshot of the software and data that were used in the research reported on in the paper An LSTM+ Model for Managing Epidemics: Using Population Mobility and Vulnerability for Forecasting COVID-19 Hospital Admissions by A. Ray, W. Jank, K. Dutta, and M. Mullarkey.

Cite

To cite this software, please cite the paper using its DOI and the software itself using the following DOI: https://doi.org/10.1287/ijoc.2023.1269.cd

Below is the BibTex for citing this version of the code.

@article{LSTMCovid,
  author =        {A. Ray, W. Jank, K. Dutta, and M. Mullarkey},
  publisher =     {INFORMS Journal on Computing},
  title =         {An LSTM+ Model for Managing Epidemics: Using Population Mobility and Vulnerability for Forecasting COVID-19 Hospital Admissions},
  year =          {2022},
  doi =           {10.1287/ijoc.2023.1269.cd},
  note =          {https://github.com/INFORMSJoC/2021.0027}
}  

Data for LSTM+ Model

This repository contains the anonymous datasets used for the case study in the LSTM+ model.

Dataset

MobilityData_Anonymized.csv: Contains the mobility metrics, ROGSI, NULV, and NUDL at the zip code level. The county and zipcode information is anonymized. SocialVulnerabilityData_Anonymized.csv: Contains the social vulnerability data at the county level. HospitalityData_Anonymized.csv: Contains the inpatient admission data for the hospitals. The hospital names are anonymized. HospitalCountyMapping_Anonymized.csv: Contains the hospital to county mapping, both data anonymized.

Simulated Dataset

sim_dynamicInput_train.csv: Contains the simulated data for the dynamic features, D. The first column indicates the time index t, whereas the subsequent column headings indicate the entity index i. This file contains only the portion of the simulated data used for training. sim_dynamicInput_train.csv: Same data as above but the holdout set for testing. sim_output_train.csv: Contains the simulated data for the output, O. The first column indicates the time index t, whereas the subsequent column headings indicate the entity index i. This file contains only the portion of the simulated data used for training. sim_output_train.csv: Same data as above but the holdout set for testing. sim_staticInput.csv: This is the static or time-invariant data, S. The first column indicates entity index i, and the second column contains the corresponding value.