Parameter Estimation and Statistical Inference

Slides:



Advertisements
Similar presentations
20-Opening 統計學 授課教師:楊維寧 10Simple-R-Commands.
Advertisements

統計學 : 應用與進階 第 11 章 : 點估計.  點估計  類比原則  最大概似法  不偏性  有效性  一致性.
第十三章 医学统计学方法的基本概念和基本步骤
第二章 语言测试的功能与分类 湖南师范大学外国语学院 邓 杰 教授.
估計的基本概念 估計量之性質 估計之方法 區間估計之基本概念 平均數之區間估計 樣本大小.
人群健康研究的统计方法 预防医学系 指导教师:方亚 电话:
概率论与数理统计 课件制作:应用数学系 概率统计课程组.
第一章 緒論.
應用統計理論 編著:劉正夫教授 Reference:1) Wonnacott and Wonnacott. Introductory
第六章 假设检验的基本概念.
Chapter 8 Liner Regression and Correlation 第八章 直线回归和相关
平均数检定 庄文忠 副教授 世新大学行政管理学系 SPSS之应用(庄文忠副教授) 2012/7/6.
Service survey center, NBS
假設檢定.
Analysis of Variance 變異數分析
Population proportion and sample proportion
型II誤差機率的計算 Calculating Type II Error Probabilities
Sampling Theory and Some Important Sampling Distributions
模式识别 Pattern Recognition
一元线性回归(二).
What are samples?. Chapter 6 Introduction to Inferential Statistics Sampling and Sampling Designs.
第十章 兩母體之假設檢定 Inferences Based on Two-Samples:
分析化学教程 第二章 分析数据处理及 分析测试的质量保证 (1) 分析化学教程( 学年)
平均数检定 庄文忠 副教授 世新大学行政管理学系 计量分析一(庄文忠副教授) 2011/7/12.
第 3 章 敘述統計:數值方法.
Continuous Probability Distributions
第 5 章 樣本資料的數值分布.
Sampling Theory and Some Important Sampling Distributions
第六講 函數極值之求法與均值定理 (Extrema & The Mean Value Theorem)
十一、簡單相關與簡單直線回歸分析(Simple Correlations and Simple Linear Regression )
簡單迴歸模型的基本假設 用最小平方法(OLS-ordinary least square)找到一個迴歸式:
教材 P.264 Point Estimation To estimate the value of a population parameter, we compute a corresponding characteristic of the sample, referred to as a sample.
第一章 敘述統計學.
Chapter 7 Sampling and Sampling Distributions
Interval Estimation區間估計
第一章.
统 计 学 (第三版) 2008 作者 贾俊平 统计学.
第 3 章 敘述統計II:數值方法 Part A (3.1~3.2).
第 7 章 抽樣與抽樣分配 Part A ( ).
The Nature and Scope of Econometrics
Workshop on Statistical Analysis
Chap 9 Testing Hypotheses and Assessing Goodness of Fit
统 计 学 (第三版) 2008 作者 贾俊平 统计学.
实验数据处理方法 王永刚.
Introduction to Basic Statistics
抽樣分配 Sampling Distributions
相關統計觀念復習 Review II.
第八章 假設之檢定與信賴區間 陳順宇 教授 成功大學統計系.
Introduction to Basic Statistics
Chapter 04 流程能力與績效分析.
以時間序列分析法偵測 台灣一等一級水準網之殘留系統誤差 Detecting Remained Systematic Errors In The First-Order ClassⅠLeveling Network of Taiwan By Using Time series 指導教授:許榮欣 學生:林曾進.
Dr. C. Hsieh College of Informatics Kao yuan University
第五章 估計與信賴區間 5.1 估計概論 估計量的分配 信賴度、信賴區間與最大容忍誤差16
Chapter 5 z-Scores.
抽样理论 与 参数估计 主讲人:孟迎芳.
The Bernoulli Distribution
Some Important Probability Distributions
商用統計學 Chapter 7 估計.
Review of Statistics.
Statistics Chapter 1 Introduction Instructor: Yanzhi Wang.
第二部分:统计推断 Chp6:统计推断概述 Chp7:非参数推断 Chp8:Bootstrap Chp9:参数推断 Chp10:假设检验
Review of Statistics.
楊志強 博士 國立台北教育大學系 教育統計學 楊志強 博士 國立台北教育大學系
品質管理與實習 : MIL-STD-105E 何正斌 國立屏東科技大學工業管理學系.
多變數函數的極值(含Lagrange法)
Multiple Regression: Estimation and Hypothesis Testing
第七章 计量资料的统计分析.
簡單迴歸分析與相關分析 莊文忠 副教授 世新大學行政管理學系 計量分析一(莊文忠副教授) 2019/8/3.
Gaussian Process Ruohua Shi Meeting
Presentation transcript:

Parameter Estimation and Statistical Inference Chapter Four Parameter Estimation and Statistical Inference

Statistics II_Chapter4 Sample and Sampling Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 抽樣方法 統計之基礎理論與觀念 簡單隨機抽樣 分層抽樣 部落抽樣 系統抽樣 Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 抽樣分配 統計之基礎理論與觀念 中央極限定理: 若母體為任意分配, 且母體之平均數為m, 變異數為s 2,則自母體抽取 n 個樣本, 若 n 夠大(n>25), 樣本平均數 樣本比例 Examples Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Central Limit Theorem Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Illustration of the Central Limit Theorem (Distribution of average scores from throwing dice) Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 例、設某產品製程是常態分配N(5,0.04),抽樣20個產品資料,試問: (1)這20個樣本平均數大於5.02的機率是多少? (2)這20個樣本平均數介於4.9到5.1的機率是多少? (3)這20個樣本總和大於100的機率是多少? (4)這20個樣本總和大於101的機率是多少? (5)這20個樣本平均數x之變異數是多少? (6)這20個樣本總和之變異數是多少? Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 (續) (1)利用中央極限定理求抽50件中樣本不良率P剛好為母體不良 率1%的機率? (2)如果重複抽樣400次,每次50個零件,請描述樣本不良率P的分 佈狀況。 (3)如果重複抽樣400次,每次100個零件,請描述不良率P分佈狀況。 Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 (1)抽樣50件,則樣本不良率P與P相差在1%以內的機率是多少? (2)抽樣500件, 則樣本不良率P與P相差在1%以內的機率是多少? (3)抽樣5000件, 則樣本不良率P與P相差在l%以內的機率是多少? Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Sample mean distribution Vs. 1. Population type 2. Sample size Horng-Chyi Horng Statistics II_Chapter4

點估計(Point Estimation) 統計之基礎理論與觀念 以抽樣得來之樣本資料, 依循某一公式計算出單一數值, 來估計母體參數, 稱為點估計. 好的點估計公式之條件: 不偏性 最小變異 常用之點估計: 母體平均數(m) 母體變異數(s 2) Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Horng-Chyi Horng Statistics II_Chapter4

Criteria for Point Estimator Unbiased Minimum Variance Absolute Efficiency Relative Efficiency Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 不偏估計式(Unbiased Estimator) Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 最小變異不偏估計式 Sample Mean X and Xi are both unbiased estimator of m, but the variance of sample mean (s2/n) is less than the variance of Xi (s2). Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 標準誤差(Standard Error) Used to measure the precision of estimation. Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Horng-Chyi Horng Statistics II_Chapter4

Absolute Efficiency 絕對有效性 Used when no unbiased estimator are available. Choose the estimator with smallest MSE. Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Relative Efficiency 相對有效性 Choose the estimator with relative smaller MSE. Horng-Chyi Horng Statistics II_Chapter4

Method of Maximum Likelihood 最大概似法 Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Horng-Chyi Horng Statistics II_Chapter4

假設檢定(Hypothesis Testing) 統計之基礎理論與觀念 “A person is innocent until proven guilty beyond a reasonable doubt.” 在沒有充分證據證明其犯罪之前, 任何人皆是清白的. 假設檢定 H0: m = 50 cm/s H1: m  50 cm/s Null Hypothesis (H0) Vs. Alternative Hypothesis (H1) One-sided and two-sided Hypotheses A statistical hypothesis is a statement about the parameters of one or more populations. Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 About Testing Critical Region Acceptance Region Critical Values Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Errors in Hypothesis Testing 統計之基礎理論與觀念 檢定結果可能為 Type I Error(a): Reject H0 while H0 is true. Type II Error(b): Fail to reject H0 while H0 is false. Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Making Conclusions 統計之基礎理論與觀念 We always know the risk of rejecting H0, i.e., a, the significant level or the risk. We therefore do not know the probability of committing a type II error (b). Two ways of making conclusion: 1. Reject H0 2. Fail to reject H0, (Do not say accept H0) or there is not enough evidence to reject H0. Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Significant Level (a) a = P(type I error) = P(reject H0 while H0 is true) n = 10, s = 2.5 s/n = 0.79 Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Horng-Chyi Horng Statistics II_Chapter4

The Power of a Statistical Test Power = 1 - b Power = the sensitivity of a statistical test Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 General Procedure for Hypothesis Testing 1. From the problem context, identify the parameter of interest. 2. State the null hypothesis, H0. 3. Specify an appropriate alternative hypothesis, H1. 4. Choose a significance level a. 5. State an appropriate test statistic. 6. State the rejection region for the statistic. 7. Compute any necessary sample quantities, substitute these into the equation for the test statistic, and compute that value. 8. Decide whether or not H0 should be rejected and report that in the problem context. Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Exercise 8-1 A Textile fiber manufacturer is investigating a new drapery yarn, which the company claims has a mean thread elongation of 12 kg with a standard deviation of 0.5kg. The company wishes to test the hypothesis H0: m = 12 against H1: m < 12, using a random sample of four specimens. What is the type I error probability if the critical region is defined as x < 11.5 kg? Find b for the case where the true mean elongation is 11.25kg. Horng-Chyi Horng Statistics II_Chapter4

Statistics II_Chapter4 Exercise 8-2, 8-3 8-2 Repeat Exercise 8-1 using a sample size of n = 16 and the same critical region. 8-3 In Exercise 8-1, find the boundary of the critical region if the type I error probability is specified to be a = 0.01. Horng-Chyi Horng Statistics II_Chapter4