Sammenlign metoder

	PatchTST	Konform forudsigelse til tidsserieprognoser
Fagområde	Dyb læring	Økonometri
Familie	Machine learning	Regression model
Oprindelsesår	2023	2021
Ophavsperson	Nie, Y. et al.	Angelopoulos & Bates (tutorial); Xu & Xie (time-series EnbPI)
Type	Transformer for time series forecasting	Distribution-free prediction interval wrapper
Oprindelig kilde	Nie, Y., Nguyen, N. H., Sinthong, P. & Kalagnanam, J. (2023). A Time Series is Worth 64 Words: Long-term Forecasting with Transformers. ICLR.	Angelopoulos, A. N. & Bates, S. (2023). Conformal Prediction: A Gentle Introduction. Foundations and Trends in Machine Learning, 16(4), 494-591.
Aliasser	PatchTST — Yama Tabanlı Zaman Serisi Transformer, patch-based time series transformer, channel-independent transformer	conformal prediction, distribution-free prediction intervals, EnbPI, Konformal Tahmin (Conformal Prediction — Zaman Serisi)
Relaterede	3	4
Resumé	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.	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).