방법 비교

	DeepAR	시계열 예측을 위한 Conformal Prediction	PatchTST
분야	딥러닝	계량경제학	딥러닝
계열	Machine learning	Regression model	Machine learning
기원 연도	2020	2021	2023
창시자	Salinas, D., Flunkert, V. & Gasthaus, J. (Amazon)	Angelopoulos & Bates (tutorial); Xu & Xie (time-series EnbPI)	Nie, Y. et al.
유형	Autoregressive recurrent neural network (probabilistic forecasting)	Distribution-free prediction interval wrapper	Transformer for time series forecasting
원전	Salinas, D., Flunkert, V., Gasthaus, J. & Januschowski, T. (2020). DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networks. International Journal of Forecasting, 36(3), 1181–1191.	Angelopoulos, A. N. & Bates, S. (2023). Conformal Prediction: A Gentle Introduction. Foundations and Trends in Machine Learning, 16(4), 494-591.	Nie, Y., Nguyen, N. H., Sinthong, P. & Kalagnanam, J. (2023). A Time Series is Worth 64 Words: Long-term Forecasting with Transformers. ICLR.
별칭	DeepAR — Olasılıksal RNN Tahmini, probabilistic autoregressive RNN forecasting, Amazon DeepAR	conformal prediction, distribution-free prediction intervals, EnbPI, Konformal Tahmin (Conformal Prediction — Zaman Serisi)	PatchTST — Yama Tabanlı Zaman Serisi Transformer, patch-based time series transformer, channel-independent transformer
관련	5	4	3
요약	DeepAR is Amazon's industrial forecasting model, introduced by Salinas, Flunkert and Gasthaus (2017; published 2020), that uses an autoregressive recurrent neural network to estimate the parameters of a probability distribution at each step, producing a confidence interval rather than a single point forecast. It can model many related time series jointly within one model.	Conformal prediction is a distribution-free wrapper that turns any point forecaster — ARIMA, a neural network, or a machine-learning model — into valid prediction intervals using only its residuals. The time-series form was popularised by Xu & Xie (2021) and the modern tutorial treatment by Angelopoulos & Bates (2023).	PatchTST is a patch-based Transformer architecture for time series forecasting, introduced by Nie and colleagues in 2023, that cuts each series into overlapping patches treated as tokens and processes channels independently. It balances computational efficiency with strong accuracy on long-horizon forecasting.