【文章內(nèi)容簡介】
變異是可加的;各處理內(nèi)及實(shí)驗(yàn)組內(nèi)部的方差一致。擴(kuò)大組間差異,減小組內(nèi)差異。 ? 回歸分析 是通過觀測(cè)值尋求自變量與因變量之間的函數(shù)關(guān)系的一種統(tǒng)計(jì)方法,它所要解決的主要問題是:在相關(guān)變量間建立數(shù)學(xué)關(guān)系式,即回歸方程;檢驗(yàn)回歸方程存在的統(tǒng)計(jì)合理性,并對(duì)各自變量對(duì)因變量影響的顯著性進(jìn)行檢驗(yàn);利用回歸方程進(jìn)行預(yù)測(cè)和控制,并了解這種結(jié)果的精確程度。 回歸分析與方差分析 單個(gè)分類變量的不同類別時(shí),等價(jià)于單因素方差分析 ,回歸系數(shù)檢驗(yàn)等價(jià)于不同類別與參照類平均值之差的 t檢驗(yàn)。 多個(gè)分類變量形成的虛擬變量 ,等價(jià)于多因素方差分析,只考慮主效應(yīng)。 既有分類變量,又有連續(xù)變量 ,等價(jià)于協(xié)方差分析,假定交互作用為 0。 (五)交互作用 當(dāng)某一自變量對(duì)因變量的作用大小與另一個(gè)自變量的取值有關(guān)時(shí),則表示兩個(gè)變量有交互作用( interaction)。 檢驗(yàn)兩變量間有無交互作用,普遍的做法是在方程中加入它們的乘積項(xiàng)再做檢驗(yàn)。如考察 X X2間的交互作用,可在模型中加入 X1X2項(xiàng)。 Interaction coefficient: C X1 and X2 must be in model for interaction to be properly specified. Interaction Model With 2 Independent Variables ? 1. Hypothesizes Interaction Between Pairs of X Variables ? Response to One X Variable Varies at Different Levels of Another X Variable Interaction Model With 2 Independent Variables ? 1. Hypothesizes Interaction Between Pairs of X Variables ? Response to One X Variable Varies at Different Levels of Another X Variable ? 2. Contains TwoWay Cross Product Terms E Y X X X Xi i i i( ) ? ? ? ?? ? ? ?0 1 1 2 2 3 1 2? 1. Hypothesizes Interaction Between Pairs of X Variables ? Response to One X Variable Varies at Different Levels of Another X Variable ? 2. Contains TwoWay Cross Product Terms ? 3. Can Be Combined With Other Models ? Example: DummyVariable Model E Y X X X Xi i i i( ) ? ? ? ?? ? ? ?0 1 1 2 2 3 1 2Interaction Model With 2 Independent Variables Effect of Interaction Effect of Interaction ? 1. Given: E Y X X X Xi i i i( ) ? ? ? ?? ? ? ?0 1 1 2 2 3 1 2Effect of Interaction ? 1. Given: ? 2. Without Interaction Term, Effect of X1 on Y Is Measured by ?1 E Y X X X Xi i i i( ) ? ? ? ?? ? ? ?0 1 1 2 2 3 1 2Effect of Interaction ? 1. Given: ? 2. Without Interaction Term, Effect of X1 on Y Is Measured by ?1 ? 3. With Interaction Term, Effect of X1 on Y Is Measured by ?1 + ?3X2 ? Effect Increases As X2i Increases E Y X X X Xi i i i( ) ? ? ? ?? ? ? ?0 1 1 2 2 3 1 2Interaction Model Relationships Interaction Model Relationships E(Y) X1 4 8 12 0 0 1 E(Y) = 1 + 2X1 + 3X2 + 4X1X2 Interaction Model Relationships E(Y) X1 4 8 12 0 0 1 E(Y) = 1 + 2X1 + 3X2 + 4X1X2 E(Y) = 1 + 2X1 + 3(0) + 4X1(0) = 1 + 2X1 Interaction Model Relationships E(Y) X1 4 8 12 0 0 1 E(Y) = 1 + 2X1 + 3X2 + 4X1X2 E(Y) = 1 + 2X1 + 3(1) + 4X1(1) = 4 + 6X1 E(Y) = 1 + 2X1 + 3(0) + 4X1(0) = 1 + 2X1 Interaction Model Relationships Effect (slope) of X1 on E(Y) does depend on X2 value E(Y) X1 4 8 12 0 0 1 E(Y) = 1 + 2X1 + 3X2 + 4X1X2 E(Y) = 1 + 2X1 + 3(1) + 4X1(1) = 4 + 6X1 E(Y) = 1 + 2X1 + 3(0) + 4X1(0) = 1 + 2X1 Interaction Model Worksheet Ca s e , i Y i X 1 i X 2 i X 1 i X 2 i1 1 1 3 32 4 8 5 403 1 3 2 64 3 5 6 30: : : : :Multiply X1 by X2 to get X1X2. Run regression with Y, X1, X2 , X1X2 路徑分析原理 ? 遞歸 (recursive)模型 Z1 Z1 Z1 Z1 e1 e2 β43 β31 β41 β32 β42 圖 1 遞歸模型路徑圖 只有單向的直線箭頭,且誤差之間沒有弧線箭頭聯(lián)系 路徑分析基本步驟 ? ,建構(gòu)一個(gè)可以檢驗(yàn)的初始模式 ,并繪出一個(gè)沒有路徑系數(shù)的路徑圖 ? ,以估計(jì)路徑系數(shù)并檢驗(yàn)其是否顯著 ,進(jìn)而估計(jì)殘差系數(shù) . ? 所謂