Emergency Department Daily Arrival Forecasts with Meteorological and Calendar Factors

5.8 KB of R and Python code implements hybrid forecasting algorithms for emergency department patient arrivals. The study by Hamed Tabesh, last updated in April 2026, incorporates meteorological and calendar influences and compares the performance of ARIMA, ANN, LSTM, GLM, and two hybrid models using RMSE, ME, and SMAPE metrics.

Use Cases

• Benchmarking hybrid ARIMA-ANN models against standalone algorithms based on the described performance metrics (RMSE, ME, SMAPE).
• Forecasting emergency department arrivals based on meteorological and calendar factors mentioned in the description.
• Developing strategic decision-making tools for managing patient flow in high-volatility periods based on the intermediate horizon findings.

Strengths

• Performance metrics for six algorithms (ARIMA, ANN, hybrid1, hybrid2, LSTM, GLM) are provided with specific RMSE, ME, and SMAPE values.
• The hybrid2 algorithm demonstrated superior performance across short, intermediate, and overall forecasting horizons.

Limitations

• Row count and column-level documentation are absent; field semantics must be inferred after download.
• The dataset is a 5.8 KB ZIP file, indicating a very limited scope, likely containing only code and not the underlying arrival data.

Provenance

Source

Hamed Tabesh via figshare

Collection Method

Likely developed as part of a research study on forecasting emergency department arrivals.

Freshness

Last updated 2026-04-30 17:47:26; freshness should be verified.