เปรียบเทียบวิธี

	PatchTST	แบบจำลอง ARIMA (Autoregressive Integrated Moving Average)	การพยากรณ์แบบคอนฟอร์มอลสำหรับอนุกรมเวลา
สาขาวิชา	การเรียนรู้เชิงลึก	เศรษฐมิติ	เศรษฐมิติ
ตระกูล	Machine learning	Regression model	Regression model
ปีกำเนิด	2023	2015	2021
ผู้ริเริ่ม	Nie, Y. et al.	Box & Jenkins (Box-Jenkins methodology)	Angelopoulos & Bates (tutorial); Xu & Xie (time-series EnbPI)
ประเภท	Transformer for time series forecasting	Univariate time-series model	Distribution-free prediction interval wrapper
แหล่งต้นตำรับ	Nie, Y., Nguyen, N. H., Sinthong, P. & Kalagnanam, J. (2023). A Time Series is Worth 64 Words: Long-term Forecasting with Transformers. ICLR.	Box, G. E. P., Jenkins, G. M., Reinsel, G. C. & Ljung, G. M. (2015). Time Series Analysis: Forecasting and Control (5th ed.). Wiley. ISBN: 978-1118675021	Angelopoulos, A. N. & Bates, S. (2023). Conformal Prediction: A Gentle Introduction. Foundations and Trends in Machine Learning, 16(4), 494-591.
ชื่อเรียกอื่น	PatchTST — Yama Tabanlı Zaman Serisi Transformer, patch-based time series transformer, channel-independent transformer	Box-Jenkins model, ARIMA(p,d,q), ARIMA Modeli	conformal prediction, distribution-free prediction intervals, EnbPI, Konformal Tahmin (Conformal Prediction — Zaman Serisi)
ที่เกี่ยวข้อง	3	5	4
สรุป	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.	ARIMA is a univariate time-series forecasting model that combines autoregressive, integrated (differencing), and moving-average components to predict a single continuous series from its own past. It is the centrepiece of the Box-Jenkins methodology set out in Box, Jenkins, Reinsel & Ljung's Time Series Analysis (5th ed., 2015).	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).