Table 12 MAPE accuracies for the proxy-only models for Clinic 1
From: Forecasting patient flows with pandemic induced concept drift using explainable machine learning
| Â | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | Overall mean | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Â | MAPE | R | MAPE | R | MAPE | R | MAPE | R | MAPE | R | MAPE | R | MAPE | R |
CatBoost | 16.9 | 4.5 | 14.0 | 4.1 | 14.0 | 4.1 | 16.5 | 3.7 | 12.8 | 3.6 | 14.0 | 3.8 | 14.8 | 4.0 |
Random Forest | 16.3 | 4.2 | 14.0 | 4.2 | 14.3 | 4.3 | 16.3 | 3.7 | 13.0 | 3.8 | 13.3 | 3.3 | 14.6 | 4.0 |
ARIMA | 9.7 | 2.0 | 9.7 | 2.5 | 9.0 | 2.3 | 13.1 | 2.6 | 10.8 | 2.7 | 10.5 | 2.4 | 10.5 | 2.4 |
Prophet | 8.5 | 1.7 | 8.4 | 1.9 | 8.4 | 2.1 | 13.3 | 2.8 | 10.2 | 2.6 | 14.2 | 3.5 | 10.2 | 2.3 |
Voting | 11.0 | 2.6 | 9.7 | 2.3 | 9.7 | 2.3 | 12.4 | 2.1 | 10.0 | 2.2 | 11.3 | 2.0 | 10.6 | 2.3 |