Subject:迴歸分析 (一) Regression Analysis (I) [110-1: 2021/09~2022/01]  (英語授課)

Instructor: Wu, Han-Ming (吳漢銘) (Associate Professor, Department of Statistics, National Chengchi University)

Office: College of Commerce, Room 261237,  Extension: 81237。

Office Hour 一/13:00~15:00E-mail: wuhm@g.nccu.edu.tw

Course Department:Commerce/B/0 。Type of Credit: Elective。Credit(s):3。

Session: Thursday 09:10-12:00, 資訊140301。(Capacity: 100人)

Prerequisite: Statistics

Hands-on course (practicum):  TBALocation TBA。TA 李其軒 (統計碩一)

Announcement

 

  • [2021/10/14] Quiz 1 on 21 Oct, 11:00~12:00。Scope: Ch1 ~ Ch2。Content: Academic Terminology, Definition, Proof, CI, Hypothesis Testing Procedure。Bring your calculator, "Smart Phone, Laptop, Tablet" are not allowed during the quiz.
  • [2021/09/23] 依【秘書處訊】實體授課規劃,9月27日起調整措施,本課程符合「 50人以下」,故以以實體授課為原則。其它重要規定: 
    • 室內人數上限80人(資訊140301容量為 100人)、維持安全社交距離(2.25平方公尺/人)。
    • 採固定座位及固定成員、落實課堂點名、全程佩戴口罩且禁止飲食、教室保持通風良好及定時清潔消毒等原則。
    • (!!重要!!) 110學年度第1學期各課程仍繼續施行「掃描QR Code點名」作業 (僅作為疫調之用)。
  • [2021/09/17] 程式加分考制度 (Bonus Test)
     - 目的: 為鼓勵同學於TA課,有好好學習利用R程式做迴歸分析
     - 型式: 自由參加 (共兩次)
     - 考試內容: 給資料(csv或xlsx檔),用R算出統計量、印出報表取值、繪圖等等。
     - 考試時間: 接著期中考、期末考後的一小時。
     - 考試範圍: 與期中考、期末考同範圍。
     - 規則: 自帶筆電&延長線,Open Book 、網路開放可查資料、禁與別人通訊。
     - 配分: 兩次各佔學期成績10%。
     - 若是線上考,則可能再另行規定一些事項。

    [2021/09/17] Bonus Test for R programming
     - Goal: to encourage students to learn R for regression analysis in the TA class
     - Type: free to take the exams  (twice)
     - Content: given csv or xlsx data file,students use R to compute ststistics, to print the report, to obtain the model, and to draw the plots。
     - Time: one hour after the midtem-exam and final exam
     - Scope: same as midtem-exam and final exam
     - Rule : bring your laptop, open book, free to use internet. Texting with others is prohibited
     - Grade: 10% for each.
  • [2021/09/16] (1) 遠距線上上課,請儘量用g.nccu的帳號登入,可不用被主持人同意,即可加入。(2) 為避免一些問題及麻煩,三次小考試皆改在正課第3堂考。
  • [2021/09/15] 第一週上課(9/16): 沒有[選點名冊]裡的同學請填點名表單!! (上課時會公告填寫開放時間)
  • [2021/09/14] (!!重要!!) 統計系辦通知,要加簽的同學請回報: 110-1: 我要加簽吳漢銘老師的「迴歸分析 (一)」
  • [2021/09/11] 本課程為「英語授課」,使用英語的情境為:  「教師教授或講解課文時全程使用英語,學生發問中英文皆可。若是說明規定、公告或注意事項,與課本內文無直接關係,則使用中文」,另TA課,如學校有特別規定,則依其規定,否則以中文進行。
  • [2021/09/10] 欲加簽本課程的同學,請列印「選課加簽單」,給老師簽名同意加簽。
    方法一:  至商學院261237室找老師簽名(請事先FB私訊,看老師是否有在辦公室)。
    方法二:  將「選課加簽單」印成pdf (不要用手機拍照或jpg檔),FB傳給老師電子簽名。
    附件: 110選課說明.pdf選課須知109.06.pdf選課Q&A_109.06.pdf
  • [2021/09/08] 本學期遠距上課使用Google Meet,固定網址為: https://meet.google.com/rts-swoe-rfc
  • [2021/09/08] 校方最新上課防疫規定(摘要):  
    一、 前兩週課程安排以遠距授課為原則
    二、 各課程防疫措施請依以下指引辦理
    (一) ​授課方式調整
    1. 選課人數80人以上室內課程以遠距授課為原則。
    2. 選課人數51~79人之室內課程以實體授課為原則,若因教室空間限制無法保持安全社交距離,考量防疫優先,各授課教師與選課同學討論後,可依課程教學需要採實體與遠距混搭方式進行。
    3. 選課人數50人以下室內課程以實體授課為原則。
    4. 本處教學發展中心已提供遠距教學相關方案與諮詢,全力協助授課教師順利進行遠距教學。
    5. 彈性採用遠距教學方式授課時,請授課教師維持教學品質並確實掌握學生出席及學習狀況。
    6. 室內/外體育課程請依體育室相關防疫規定辦理。
    7. 實習課應採固定分組,並避免學生共用設備、器材;如有輪替使用設備或器材之需要,輪替前應先澈底消毒。
    (二) 實體授課防疫措施
    1. 各課程若經師生討論取得共識後採實體授課,實體授課時教室內師生人數以80人為上限,並須維持安全社交距離(2.25平方米/人),若因教室空間限制,各授課教師與選課同學討論後,可依課程教學需要採實體與遠距混搭方式進行,以符合安全社交距離規範。
    2. 各授課教師如對110學年度第1學期課程教室安排有疫情上之疑慮(例如通風、修課同學密集等),敬請開課單位儘速洽詢本處課務組更換教室,本處課務組將就教室容量與安排現況盡力協助更換。
    3. 師生進入室內空間時請落實手部消毒,課程進行期間請注意教室通風,並請全程佩戴口罩和禁止飲食。
    4. 因應新冠肺炎疫調需求,110學年度第1學期各課程仍繼續施行「掃描QR Code點名」作業,敬請師生實體上課時務必配合實施,以落實實聯制並完備疫調資訊。
    5. 本處再次重申QR Code點名紀錄僅作為疫調之用,且因授課方式已彈性調整,QR Code點名紀錄無法確實呈現選課同學出席狀況,敬請授課教師勿將QR Code點名紀錄作為評定學生出席率成績之參考。
  • [2021/09/08] 課務組公告110學年度第1學期本校各課程防疫指引  (2021/09/08)
  • [2021/09/08] (!!重要!!) 請修課同學加入FB Messenger課程聊天室: 「110-1-迴歸分析 (一)」。
    (加入方法: (1) 已在聊天室之同學可將未加入的同學加入,或(2) 同學們FB私訊老師(請註明課名),由老師幫忙加入。 )
  • [2021/08/05] Download the handouts and exercises below.
  • [2021/08/05] Teaching plan。Note that the 「Tentative Syllabus」is subject to change depending on class progress and other factors.

 

Course Description

A linear regression model is a relationship between an outcome and a set of predictors of interest based on the linear assumptions. It is the most important statistical analysis tool for data scientists. This course introduces the fundamental theories, methods and practical application skills in regression analysis and their generalizations. The textbook used in this course is "Michael H. Kutner et al. (2019), Applied Linear Statistical Models: Applied Linear Regression Models, Mcgraw-Hill Inc., (5th edition)". The topics in this semester cover the simple linear regression, multiple regression, inferences, model diagnostics and remedial measures, regression models for quantitative and qualitative predictors and logistic regression. In addition, students will learn how to use R/RStudio to perform the real data analysis and interpret the results. Note that the main teaching method in this class is lecturing in English. (The "Course Schedule & Requirements" below is subject to change according to the actual progress of the class.)

Course Objectives & Learning Outcomes

After completing this course, students will be able to (1) understand the basic mathematical concepts and principles of the linear regression models and their limitations; (2) evaluate and diagnose the regression models; (3) apply corrections to some real data problems in regression; (4) conduct the analysis to develop an optimal regression model using R/RStudio software.

 

Tentative Syllabus (the syllabus is always subject to change according to the needs of the course as the professor sees fit):

Week Month/Day Topics

Notes

1 09/16  [遠距] Course Introduction, Ch 1: Simple Linear Regression (SLR)
2 09/23  [遠距]  Ch2: Inferences in Regression
3 09/30 [實體]
掃QR Code
Ch2: Correlation Analysis
4 10/07[實體]
掃QR Code
Ch3: Model Diagnostics
5 10/14[實體]
掃QR Code
Ch3: Remedial Measure quiz (1)
6 10/21
Ch4: Simultaneous Inferences quiz (1) 
7 10/28
Ch5: Matrix Approach to SLR  
8 11/04 Case studies  (I), Exercise using R (I)
9 11/11

Mid-term Exam: Ch1-Ch5

Midterm Exam

10 11/18 Ch6: Multiple Regression (I)
11 11/25 Ch7: Multiple Regression (II)
12 12/02 Ch8: Regression Models for Quantitative and Qualitative Predictors quiz (2)
13 12/09 Ch9: Model Selection and Validation  
14 12/16 Ch10: Model Diagnostics
15 12/23
Ch11: Model Remedial Measures

quiz (3)

16 12/30 Ch14: Logistic Regression (Optional)  
17 01/06
Case studies (II), Exercise using R (II)
18 01/13
Final Exam: Ch6-Ch11 (Ch14) Final Exam

Textbook: Michael H. Kutner et al. (2019), Applied Linear Statistical Models: Applied Linear Regression Models, Mcgraw-Hill Inc., (5th edition)
(購買方式: (1) 華泰文化。(2) 巨流政大書城)

Michael H. Kutner et al. (2019), Applied Linear Statistical Models: Applied Linear Regression Models, Mcgraw-Hill Inc., (5th edition)( 華泰文化)

Reference


Grading Scheme:(調整配分需經
全班大多數修課同學同意)

  • Quizzes:30 % (Three quizzes, each 10%)。
  • Midtem exam:30 %。
  • Final exam:40 %
  • TA 0%。
  • HW 0%。
  • Attendance (including TA class) extra 10%。
  • EXtra (up to 10%): in-class performance/discussion, learning attitude, and so on。(No adjustment made for personal reasons)。
     

Notes (in class)

  • The lecture is based on the use of projector and handwriting tablet. Please print the lecture notes before class.
  • Rules on leave-taking by students. (缺課、曠課相關規定,依校規辦理)。
  • Treat each other with mutual respect in the classroom. (上課以「互相尊重」為最高原則並盡到「告知老師」的義務。)
  •  What you can do in the class: (1) whispered discussion, (2) go to toilet quietly, (3) eating and drinking (without alcohol) but keeping the classroom clean, (4) use laptop or tablet to take notes or pitcures.
  • What you can't do in the class: (1) play cell phone or tablet (please mute the phone), (2) chat, sleep, play cards, smoke。
  • If you have any questions, please contact TA or Lecturer directly or using e-mail or FB

 

Notes (quizzes、grading)

  • The time for the quiz is scheduled at TA class. There will be about 3~4 questions。
  • The make-up quiz/exam is not allowed for no particular reason. Only one make-up is limited  out of 3 quizzes.
  • Cheating in exams is absolutely prohibited. Any form of cheating on an exam will result in a 0 for all the rest exams.
  • The scores for the attendance is extra (which is not contained in 100%).
  • The students should attend the classes at least 2/3 of all classes during all the semester so that he/she could get these extra scores.
    (對成績有疑問,請於當次成績公佈後一星期內連絡老師。)