Multivariate Time Series Imputation On Uci
Metrics
MAE (10% missing)
Results
Performance results of various models on this benchmark
Model Name | MAE (10% missing) | Paper Title | Repository |
|---|---|---|---|
| M-RNN | 0.248 | Estimating Missing Data in Temporal Data Streams Using Multi-directional Recurrent Neural Networks | |
| BRITS | 0.219 | BRITS: Bidirectional Recurrent Imputation for Time Series | |
| ImputeTS | 0.363 | imputeTS: Time Series Missing Value Imputation in R | |
| Matrix Factorization (MF) | 0.879 | - | - |
| MICE | 0.477 | Multiple imputation using chained equations: issues and guidance for practice |
0 of 5 row(s) selected.