Time Series Forecasting on ETTh1 -> ETTm2 (test)
0.189MSETIME-LLM
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| TIME-LLMHorizon=96, Evaluation Protocol=Zero-shot2023.10 | 0.189 | 0.293 | |
| PatchTSTHorizon=96, Evaluation Protocol=Zero-shot2023.10 | 0.215 | 0.304 | |
| GPT4TSHorizon=96, Evaluation Protocol=Zero-shot2023.10 | 0.236 | 0.315 | |
| TIME-LLMHorizon=192, Evaluation Protocol=Zero-shot2023.10 | 0.237 | 0.312 | |
| TimesNetHorizon=96, Evaluation Protocol=Zero-shot2023.10 | 0.239 | 0.313 | |
| DLinearHorizon=96, Evaluation Protocol=Zero-shot2023.10 | 0.255 | 0.357 | |
| TIME-LLMHorizon=Avg, Evaluation Protocol=Zero-shot2023.10 | 0.273 | 0.34 | |
| PatchTSTHorizon=192, Evaluation Protocol=Zero-shot2023.10 | 0.275 | 0.339 | |
| GPT4TSHorizon=192, Evaluation Protocol=Zero-shot2023.10 | 0.287 | 0.342 | |
| TimesNetHorizon=192, Evaluation Protocol=Zero-shot2023.10 | 0.291 | 0.342 | |
| TIME-LLMHorizon=336, Evaluation Protocol=Zero-shot2023.10 | 0.291 | 0.365 | |
| STELLAevaluation_mode=zero-shot, training_source=ETTh12025.12 | 0.312 | 0.353 | |
| PatchTSTHorizon=Avg, Evaluation Protocol=Zero-shot2023.10 | 0.314 | 0.36 | |
| TiDEevaluation_mode=zero-shot, training_source=ETTh12025.12 | 0.314 | 0.355 | |
| TimeLLMevaluation_mode=zero-shot, training_source=ETTh12025.12 | 0.315 | 0.357 | |
| GPT4TSevaluation_mode=zero-shot, training_source=ETTh12025.12 | 0.317 | 0.359 | |
| PatchTSTevaluation_mode=zero-shot, training_source=ETTh12025.12 | 0.318 | 0.362 | |
| GPT4TSHorizon=Avg, Evaluation Protocol=Zero-shot2023.10 | 0.325 | 0.363 | |
| TimesNetHorizon=Avg, Evaluation Protocol=Zero-shot2023.10 | 0.327 | 0.361 | |
| PatchTSTHorizon=336, Evaluation Protocol=Zero-shot2023.10 | 0.334 | 0.373 | |
| DLinearHorizon=192, Evaluation Protocol=Zero-shot2023.10 | 0.338 | 0.413 | |
| TimesNetevaluation_mode=zero-shot, training_source=ETTh12025.12 | 0.339 | 0.37 | |
| GPT4TSHorizon=336, Evaluation Protocol=Zero-shot2023.10 | 0.341 | 0.374 | |
| TimesNetHorizon=336, Evaluation Protocol=Zero-shot2023.10 | 0.342 | 0.371 | |
| AutoformerHorizon=96, Evaluation Protocol=Zero-shot2023.10 | 0.352 | 0.432 | |
| FEDformerevaluation_mode=zero-shot, training_source=ETTh12025.12 | 0.357 | 0.403 | |
| TIME-LLMHorizon=720, Evaluation Protocol=Zero-shot2023.10 | 0.372 | 0.39 | |
| DLinearevaluation_mode=zero-shot, training_source=ETTh12025.12 | 0.399 | 0.439 | |
| AutoformerHorizon=192, Evaluation Protocol=Zero-shot2023.10 | 0.413 | 0.46 | |
| DLinearHorizon=Avg, Evaluation Protocol=Zero-shot2023.10 | 0.415 | 0.452 | |
| DLinearHorizon=336, Evaluation Protocol=Zero-shot2023.10 | 0.425 | 0.465 | |
| PatchTSTHorizon=720, Evaluation Protocol=Zero-shot2023.10 | 0.431 | 0.424 | |
| TimesNetHorizon=720, Evaluation Protocol=Zero-shot2023.10 | 0.434 | 0.419 | |
| GPT4TSHorizon=720, Evaluation Protocol=Zero-shot2023.10 | 0.435 | 0.422 | |
| AutoformerHorizon=Avg, Evaluation Protocol=Zero-shot2023.10 | 0.457 | 0.483 | |
| AutoformerHorizon=336, Evaluation Protocol=Zero-shot2023.10 | 0.465 | 0.489 | |
| AutoformerHorizon=720, Evaluation Protocol=Zero-shot2023.10 | 0.599 | 0.551 | |
| DLinearHorizon=720, Evaluation Protocol=Zero-shot2023.10 | 0.64 | 0.573 | |
| LLMTimeHorizon=96, Evaluation Protocol=Zero-shot2023.10 | 0.646 | 0.563 | |
| LLMTimeHorizon=192, Evaluation Protocol=Zero-shot2023.10 | 0.934 | 0.654 | |
| Crossformerevaluation_mode=zero-shot, training_source=ETTh12025.12 | 1.12 | 0.789 | |
| LLMTimeHorizon=336, Evaluation Protocol=Zero-shot2023.10 | 1.157 | 0.728 | |
| LLMTimeHorizon=Avg, Evaluation Protocol=Zero-shot2023.10 | 1.867 | 0.869 | |
| MICNevaluation_mode=zero-shot, training_source=ETTh12025.12 | 2.428 | 1.236 | |
| LLMTimeHorizon=720, Evaluation Protocol=Zero-shot2023.10 | 4.73 | 1.531 |