114年第2學期-1595 時間序列 課程資訊

評分方式

評分項目	配分比例	說明
平時成績	40	包含出席、課堂作業
期中作業	20
期末報告	40

教育目標

本課程旨在引導學生深入了解時間序列資料的特性，並掌握處理與分析時間序列資料的核心統計方法。透過結合理論學習、軟體應用及實務研究，課程目標包括： 1. 建立時間序列基礎概念： 使學生理解時間序列資料（如季節性、趨勢、循環等）的獨特性與分析的重要性。 2. 掌握多元分析技術： 訓練學生熟練運用迴歸分析、分解法、指數平滑法及 Box-Jenkins (ARIMA) 方法等經典模型進行建模與分析。 3. 強化實務應用能力： 培養學生運用專業統計軟體（主要以R語言）處理真實世界的資料，進行有效的預測與評估模型表現。

課程概述

A time series is a sequence of observations that are arranged according to the time of their outcome. The reasons of doing time series analysis are diverse, depending on the background of applications. Statisticians usually view a time series as a realization from a stochastic process. A fundamental task is to unveil the probability law that governs the observed time series. For instance, we wish to gain a better understanding of the data generating mechanism, the prediction of future values. Many recent applications of time series receive much of attention in financial areas. However, time series analysis has also exhibited its importance across many scientific areas for a long history and the desire of such analysis is still going on. In this course, our flow is to spend around two thirds of time in the treatment from a traditional look of time series then to some recent models such as GARCH, long memory models and nonparametric methods. The course level is set for master students.

基本資料

選修課，學分數：0-3
上課時間：一/2,3,4[M442]
修課班級：統計系2-4
修課年級：2年級以上
選課備註：大數據資料群組(110-114適用)，第一堂未出席者視同放棄。

參考書目

Bowerman, O’Connell, and Koehler (2005) Forecasting, Time Series, and Regression, 4th edition

Hyndman, R.~J. & Athanasopoulos, G. (2021) Forecasting: principles and practice, 3rd edition, OTexts.org/fpp3/