<font face="NimbusRomNo9L-Medi" size="5"><font face="NimbusRomNo9L-Medi" size="5"><p align="left">Contents</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">Preface xi</p><p align="left">1 Overview of the General Linear Model 1</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1</p><p align="left">1.2 The General Linear Model . . . . . . . . . . . . . . . . . . . . . . 1</p><p align="left">1.3 The Restricted General Linear Model . . . . . . . . . . . . . . . . 3</p><p align="left">1.4 The Multivariate Normal Distribution . . . . . . . . . . . . . . . . 4</p><p align="left">1.5 Elementary Properties of Normal Random Variables . . . . . . . . . 8</p><p align="left">1.6 Hypothesis Testing . . . . . . . . . . . . . . . . . . . . . . . . . . 9</p><p align="left">1.7 Generating Multivariate Normal Data . . . . . . . . . . . . . . . . 10</p><p align="left">1.8 Assessing Univariate Normality . . . . . . . . . . . . . . . . . . . 11</p><p align="left">1.8.1 Normally and Nonnormally Distributed Data . . . . . . . . 12</p><p align="left">1.8.2 Real Data Example . . . . . . . . . . . . . . . . . . . . . . 15</p><p align="left">1.9 Assessing Multivariate Normality with Chi-square Plots . . . . . . . 15</p><p align="left">1.9.1 Multivariate Normal Data . . . . . . . . . . . . . . . . . . 18</p><p align="left">1.9.2 Real Data Example . . . . . . . . . . . . . . . . . . . . . . 19</p><p align="left">1.10 Using SAS INSIGHT . . . . . . . . . . . . . . . . . . . . . . . . . 19</p><p align="left">1.10.1 Ramus Bone Data . . . . . . . . . . . . . . . . . . . . . . 19</p><p align="left">1.10.2 Risk-taking Behavior Data . . . . . . . . . . . . . . . . . . 21</p><p align="left">1.11 Three-Dimensional Plots . . . . . . . . . . . . . . . . . . . . . . . 23</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">2 Unrestricted General Linear Models 25</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25</p><p align="left">2.2 Linear Models without Restrictions . . . . . . . . . . . . . . . . . . 25</p><p align="left">2.3 Hypothesis Testing . . . . . . . . . . . . . . . . . . . . . . . . . . 26</p><p align="left">2.4 Simultaneous Inference . . . . . . . . . . . . . . . . . . . . . . . . 28</p><p align="left">2.5 Multiple Linear Regression . . . . . . . . . . . . . . . . . . . . . . 30</p><p align="left">2.5.1 Classical and Normal Regression Models . . . . . . . . . . 31</p><p align="left">2.5.2 Random Classical and Jointly Normal Regression Models . 42</p><p align="left">2.6 Linear Mixed Models . . . . . . . . . . . . . . . . . . . . . . . . . 49</p><p>2.7 One-Way Analysis of Variance . . . . . . . . . . . . . . . . . . . . 53</p><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">2.7.1 Unrestricted Full Rank One-way Design . . . . . . . . . . . 54</p><p align="left">2.7.2 Simultaneous Inference for the One-Way Design . . . . . . 56</p><p align="left">2.7.3 Multiple Testing . . . . . . . . . . . . . . . . . . . . . . . 58</p><p align="left">2.8 Multiple Linear Regression:Calibration . . . . . . . . . . . . . . . 58</p><p align="left">2.8.1 Multiple Linear Regression: Prediction . . . . . . . . . . . 68</p><p align="left">2.9 Two-way Nested Designs . . . . . . . . . . . . . . . . . . . . . . . 70</p><p align="left">2.10 Intraclass Covariance Models . . . . . . . . . . . . . . . . . . . . . 72</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">3 Restricted General Linear Models 77</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77</p><p align="left">3.2 Estimation and Hypothesis Testing . . . . . . . . . . . . . . . . . . 77</p><p align="left">3.3 Two-Way Factorial Design without Interaction . . . . . . . . . . . . 79</p><p align="left">3.4 Latin Square Designs . . . . . . . . . . . . . . . . . . . . . . . . . 87</p><p align="left">3.5 Repeated Measures Designs . . . . . . . . . . . . . . . . . . . . . 89</p><p align="left">3.5.1 Univariate Mixed ANOVA Model, Full Rank Representation</p><p align="left">for a Split Plot Design . . . . . . . . . . . . . . . . . . . . 90</p><p align="left">3.5.2 Univariate Mixed Linear Model, Less than Full Rank Representation</p><p align="left">. . . . . . . . . . . . . . . . . . . . . . . . . . . 95</p><p align="left">3.5.3 Test for Equal Covariance Matrices and for Circularity . . . 97</p><p align="left">3.6 Analysis of Covariance . . . . . . . . . . . . . . . . . . . . . . . . 100</p><p align="left">3.6.1 ANCOVA with One Covariate . . . . . . . . . . . . . . . . 102</p><p align="left">3.6.2 ANCOVA with Two Covariates . . . . . . . . . . . . . . . 104</p><p align="left">3.6.3 ANCOVA Nested Designs . . . . . . . . . . . . . . . . . . 106</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">4 Weighted General Linear Models 109</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109</p><p align="left">4.2 Estimation and Hypothesis Testing . . . . . . . . . . . . . . . . . . 110</p><p align="left">4.3 OLSE versus FGLS . . . . . . . . . . . . . . . . . . . . . . . . . . 113</p><p align="left">4.4 General Linear Mixed Model Continued . . . . . . . . . . . . . . . 114</p><p align="left">4.4.1 Example: Repeated Measures Design . . . . . . . . . . . . 117</p><p align="left">4.4.2 Estimating the <font face="CMMI10" size="2"><font face="CMMI10" size="2">df </font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">for the </font></font><font face="CMMI10" size="2"><font face="CMMI10" size="2">F </font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">statistic in GLMMs . . . . . . . 118<p align="left">4.5 Maximum Likelihood Estimation and Fisher’s Information Matrix . 119</p><p align="left">4.6 WLSE for data Heteroscedasticity . . . . . . . . . . . . . . . . . . 121</p><p align="left">4.7 WLSE for Correlated Errors . . . . . . . . . . . . . . . . . . . . . 124</p><p align="left">4.8 FGLS for Categorical Data . . . . . . . . . . . . . . . . . . . . . . 127</p><p align="left">4.8.1 Overview of the Categorical Data Model . . . . . . . . . . 127</p><p align="left">4.8.2 Marginal Homogeneity . . . . . . . . . . . . . . . . . . . . 130</p><p align="left">4.8.3 Homogeneity of Proportions . . . . . . . . . . . . . . . . . 132</p><p align="left">4.8.4 Independence . . . . . . . . . . . . . . . . . . . . . . . . . 138</p><p align="left">4.8.5 Univariate Mixed Linear Model, Less than Full Rank Representation</p><p>. . . . . . . . . . . . . . . . . . . . . . . . . . . 141</p><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">5 Multivariate General Linear Models 143</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143</p><p align="left">5.2 Developing the Model . . . . . . . . . . . . . . . . . . . . . . . . 143</p><p align="left">5.3 Estimation Theory and Hypothesis Testing . . . . . . . . . . . . . . 145</p><p align="left">5.4 Multivariate Regression . . . . . . . . . . . . . . . . . . . . . . . . 152</p><p align="left">5.5 Classical and Normal Multivariate Linear Regression Models . . . . 153</p><p align="left">5.6 Jointly Multivariate Normal Regression Model . . . . . . . . . . . 163</p><p align="left">5.7 Multivariate Mixed Models and the Analysis of Repeated Measurements</p><p align="left">. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171</p><p align="left">5.8 Extended Linear Hypotheses . . . . . . . . . . . . . . . . . . . . . 176</p><p align="left">5.9 Multivariate Regression: Calibration and Prediction . . . . . . . . . 182</p><p align="left">5.9.1 Fixed <font face="CMMI10" size="2"><font face="CMMI10" size="2">X </font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">. . . . . . . . . . . . . . . . . . . . . . . . . . . . 182<p align="left">5.9.2 Random <font face="CMMI10" size="2"><font face="CMMI10" size="2">X </font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">. . . . . . . . . . . . . . . . . . . . . . . . . . 185<p align="left">5.9.3 Random <font face="CMMI10" size="2"><font face="CMMI10" size="2">X</font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">, Prediction . . . . . . . . . . . . . . . . . . . . 186<p align="left">5.9.4 Overview - Candidate Model . . . . . . . . . . . . . . . . . 186</p><p align="left">5.9.5 Prediction and Shrinkage . . . . . . . . . . . . . . . . . . . 187</p><p align="left">5.10 Multivariate Regression: Influential Observations . . . . . . . . . . 189</p><p align="left">5.10.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 191</p><p align="left">5.11 Nonorthogonal MANOVA designs . . . . . . . . . . . . . . . . . . 192</p><p align="left">5.11.1 Unweighted Analysis . . . . . . . . . . . . . . . . . . . . . 197</p><p align="left">5.11.2 Weighted Analysis . . . . . . . . . . . . . . . . . . . . . . 198</p><p align="left">5.12 MANCOVA Designs . . . . . . . . . . . . . . . . . . . . . . . . . 200</p><p align="left">5.12.1 Overall tests . . . . . . . . . . . . . . . . . . . . . . . . . 200</p><p align="left">5.12.2 Tests of Additional Information . . . . . . . . . . . . . . . 203</p><p align="left">5.12.3 Results and Interpretation . . . . . . . . . . . . . . . . . . 204</p><p align="left">5.13 Stepdown Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 206</p><p align="left">5.14 Repeated Measures Analysis . . . . . . . . . . . . . . . . . . . . . 208</p><p align="left">5.14.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 209</p><p align="left">5.15 Extended Linear Hypotheses . . . . . . . . . . . . . . . . . . . . . 216</p><p align="left">5.15.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 219</p></font></font></p></font></font></p></font></font></p></font></font><font face="CMMI10" size="2"><font face="CMMI10" size="2">X </font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">. . . . . . . . . . . . . . . . . . . . . . . . . . . . 182<p align="left">5.9.2 Random <font face="CMMI10" size="2"><font face="CMMI10" size="2">X </font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">. . . . . . . . . . . . . . . . . . . . . . . . . . 185<p align="left">5.9.3 Random <font face="CMMI10" size="2"><font face="CMMI10" size="2">X</font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">, Prediction . . . . . . . . . . . . . . . . . . . . 186<p align="left">5.9.4 Overview - Candidate Model . . . . . . . . . . . . . . . . . 186</p><p align="left">5.9.5 Prediction and Shrinkage . . . . . . . . . . . . . . . . . . . 187</p><p align="left">5.10 Multivariate Regression: Influential Observations . . . . . . . . . . 189</p><p align="left">5.10.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 191</p><p align="left">5.11 Nonorthogonal MANOVA designs . . . . . . . . . . . . . . . . . . 192</p><p align="left">5.11.1 Unweighted Analysis . . . . . . . . . . . . . . . . . . . . . 197</p><p align="left">5.11.2 Weighted Analysis . . . . . . . . . . . . . . . . . . . . . . 198</p><p align="left">5.12 MANCOVA Designs . . . . . . . . . . . . . . . . . . . . . . . . . 200</p><p align="left">5.12.1 Overall tests . . . . . . . . . . . . . . . . . . . . . . . . . 200</p><p align="left">5.12.2 Tests of Additional Information . . . . . . . . . . . . . . . 203</p><p align="left">5.12.3 Results and Interpretation . . . . . . . . . . . . . . . . . . 204</p><p align="left">5.13 Stepdown Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 206</p><p align="left">5.14 Repeated Measures Analysis . . . . . . . . . . . . . . . . . . . . . 208</p><p align="left">5.14.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 209</p><p align="left">5.15 Extended Linear Hypotheses . . . . . . . . . . . . . . . . . . . . . 216</p><p align="left">5.15.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 219</p></font></font></p></font></font></p></font></font><font face="CMMI10" size="2"><font face="CMMI10" size="2">X </font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">. . . . . . . . . . . . . . . . . . . . . . . . . . 185<p align="left">5.9.3 Random <font face="CMMI10" size="2"><font face="CMMI10" size="2">X</font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">, Prediction . . . . . . . . . . . . . . . . . . . . 186<p align="left">5.9.4 Overview - Candidate Model . . . . . . . . . . . . . . . . . 186</p><p align="left">5.9.5 Prediction and Shrinkage . . . . . . . . . . . . . . . . . . . 187</p><p align="left">5.10 Multivariate Regression: Influential Observations . . . . . . . . . . 189</p><p align="left">5.10.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 191</p><p align="left">5.11 Nonorthogonal MANOVA designs . . . . . . . . . . . . . . . . . . 192</p><p align="left">5.11.1 Unweighted Analysis . . . . . . . . . . . . . . . . . . . . . 197</p><p align="left">5.11.2 Weighted Analysis . . . . . . . . . . . . . . . . . . . . . . 198</p><p align="left">5.12 MANCOVA Designs . . . . . . . . . . . . . . . . . . . . . . . . . 200</p><p align="left">5.12.1 Overall tests . . . . . . . . . . . . . . . . . . . . . . . . . 200</p><p align="left">5.12.2 Tests of Additional Information . . . . . . . . . . . . . . . 203</p><p align="left">5.12.3 Results and Interpretation . . . . . . . . . . . . . . . . . . 204</p><p align="left">5.13 Stepdown Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 206</p><p align="left">5.14 Repeated Measures Analysis . . . . . . . . . . . . . . . . . . . . . 208</p><p align="left">5.14.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 209</p><p align="left">5.15 Extended Linear Hypotheses . . . . . . . . . . . . . . . . . . . . . 216</p><p align="left">5.15.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 219</p></font></font></p></font></font><font face="CMMI10" size="2"><font face="CMMI10" size="2">X</font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">, Prediction . . . . . . . . . . . . . . . . . . . . 186<p align="left">5.9.4 Overview - Candidate Model . . . . . . . . . . . . . . . . . 186</p><p align="left">5.9.5 Prediction and Shrinkage . . . . . . . . . . . . . . . . . . . 187</p><p align="left">5.10 Multivariate Regression: Influential Observations . . . . . . . . . . 189</p><p align="left">5.10.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 191</p><p align="left">5.11 Nonorthogonal MANOVA designs . . . . . . . . . . . . . . . . . . 192</p><p align="left">5.11.1 Unweighted Analysis . . . . . . . . . . . . . . . . . . . . . 197</p><p align="left">5.11.2 Weighted Analysis . . . . . . . . . . . . . . . . . . . . . . 198</p><p align="left">5.12 MANCOVA Designs . . . . . . . . . . . . . . . . . . . . . . . . . 200</p><p align="left">5.12.1 Overall tests . . . . . . . . . . . . . . . . . . . . . . . . . 200</p><p align="left">5.12.2 Tests of Additional Information . . . . . . . . . . . . . . . 203</p><p align="left">5.12.3 Results and Interpretation . . . . . . . . . . . . . . . . . . 204</p><p align="left">5.13 Stepdown Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 206</p><p align="left">5.14 Repeated Measures Analysis . . . . . . . . . . . . . . . . . . . . . 208</p><p align="left">5.14.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 209</p><p align="left">5.15 Extended Linear Hypotheses . . . . . . . . . . . . . . . . . . . . . 216</p><p align="left">5.15.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 219</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">6 Doubly Multivariate Linear Model 223</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223</p><p align="left">6.2 Classical Model Development . . . . . . . . . . . . . . . . . . . . 223</p><p align="left">6.3 Responsewise Model Development . . . . . . . . . . . . . . . . . . 226</p><p align="left">6.4 The Multivariate Mixed Model . . . . . . . . . . . . . . . . . . . . 227</p><p align="left">6.5 Double Multivariate and Mixed Models . . . . . . . . . . . . . . . 231</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">7 The Restricted MGLM and the Growth Curve Model 243</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243</p><p align="left">7.2 The Restricted Multivariate General Linear Model . . . . . . . . . . 243</p><p align="left">7.3 The GMANOVA Model . . . . . . . . . . . . . . . . . . . . . . . 247</p><p align="left">7.4 Canonical Form of the GMANOVA Model . . . . . . . . . . . . . . 253</p><p align="left">7.5 Restricted Nonorthogonal Three-Factor Factorial MANOVA . . . . 259</p><p>7.5.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 269</p><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">7.6 Restricted Intraclass Covariance Design . . . . . . . . . . . . . . . 269</p><p align="left">7.6.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 275</p><p align="left">7.7 Growth Curve Analysis . . . . . . . . . . . . . . . . . . . . . . . . 279</p><p align="left">7.7.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 283</p><p align="left">7.8 Multiple Response Growth Curves . . . . . . . . . . . . . . . . . . 289</p><p align="left">7.8.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 290</p><p align="left">7.9 Single Growth Curve . . . . . . . . . . . . . . . . . . . . . . . . . 294</p><p align="left">7.9.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 294</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">8 The SUR Model and the Restricted GMANOVA model 297</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297</p><p align="left">8.2 The MANOVA-GMANOVA Model . . . . . . . . . . . . . . . . . 297</p><p align="left">8.3 Tests of Fit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303</p><p align="left">8.4 Sum of Profiles and CGMANOVA Models . . . . . . . . . . . . . . 305</p><p align="left">8.5 The SUR Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307</p><p align="left">8.6 The Restricted GMANOVA Model . . . . . . . . . . . . . . . . . . 314</p><p align="left">8.7 GMANOVA-SUR: One Population . . . . . . . . . . . . . . . . . . 317</p><p align="left">8.7.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 317</p><p align="left">8.8 GMANOVA-SUR: Several Populations . . . . . . . . . . . . . . . 319</p><p align="left">8.8.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 319</p><p align="left">8.9 SUR Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319</p><p align="left">8.9.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 323</p><p align="left">8.10 Two-Period Crossover Design with Changing Covariates . . . . . . 328</p><p align="left">8.10.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 329</p><p align="left">8.11 Repeated Measurements with Changing Covariates . . . . . . . . . 334</p><p align="left">8.11.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 335</p><p align="left">8.12 MANOVA-GMANOVA Model . . . . . . . . . . . . . . . . . . . . 337</p><p align="left">8.12.1 Results and interpretation . . . . . . . . . . . . . . . . . . 338</p><p align="left">8.13 CGMANOVA Model . . . . . . . . . . . . . . . . . . . . . . . . . 344</p><p align="left">8.13.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 346</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">9 Simultaneous Inference Using Finite Intersection Tests 349</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349</p><p align="left">9.2 Finite Intersection Tests . . . . . . . . . . . . . . . . . . . . . . . . 349</p><p align="left">9.3 Finite Intersection Tests of Univariate Means . . . . . . . . . . . . 350</p><p align="left">9.4 Finite Intersection Tests for Linear Models . . . . . . . . . . . . . . 354</p><p align="left">9.5 A Comparisons of Some Tests of Univariate Means . . . . . . . . . 355</p><p align="left">9.5.1 Single-Step Methods . . . . . . . . . . . . . . . . . . . . . 355</p><p align="left">9.5.2 Stepdown Methods . . . . . . . . . . . . . . . . . . . . . . 357</p><p align="left">9.6 Analysis of Means Analysis . . . . . . . . . . . . . . . . . . . . . 358</p><p align="left">9.7 Simultaneous Test Procedures for Mean Vectors . . . . . . . . . . . 360</p><p align="left">9.8 Finite Intersection Test of Mean Vectors . . . . . . . . . . . . . . . 362</p><p align="left">9.9 Finite Intersection Test of Mean Vectors with Covariates . . . . . . 366</p><p align="left">9.10 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368</p><p>9.11 Univariate: One-way ANOVA . . . . . . . . . . . . . . . . . . . . 369</p><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">9.12 Multivariate: One-way MANOVA . . . . . . . . . . . . . . . . . . 372</p><p align="left">9.13 Multivariate: One-way MANCOVA . . . . . . . . . . . . . . . . . 379</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">10 Computing Power for Univariate and Multivariate GLM 381</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381</p><p align="left">10.2 Power for Univariate GLMs . . . . . . . . . . . . . . . . . . . . . 383</p><p align="left">10.3 Estimating Power, Sample Size, and Effect Size for the GLM . . . . 384</p><p align="left">10.3.1 Power and Sample Size . . . . . . . . . . . . . . . . . . . . 384</p><p align="left">10.3.2 Effect Size . . . . . . . . . . . . . . . . . . . . . . . . . . 385</p><p align="left">10.4 Power and Sample Size based upon Interval-Estimation . . . . . . . 388</p><p align="left">10.5 Calculating Power and Sample Size for Some Mixed Models . . . . 390</p><p align="left">10.5.1 Random One-Way ANOVA Design . . . . . . . . . . . . . 390</p><p align="left">10.5.2 Two Factor Mixed Nested ANOVA Design . . . . . . . . . 396</p><p align="left">10.6 Power for Multivariate GLMs . . . . . . . . . . . . . . . . . . . . 400</p><p align="left">10.7 Power and Effect Size Analysis for Univariate GLMs . . . . . . . . 401</p><p align="left">10.7.1 One-Way ANOVA . . . . . . . . . . . . . . . . . . . . . . 401</p><p align="left">10.7.2 Three-Way ANOVA . . . . . . . . . . . . . . . . . . . . . 403</p><p align="left">10.7.3 One-Way ANCOVA Design with two covariates . . . . . . 405</p><p align="left">10.8 Power and Sample Size based upon Interval-Estimation . . . . . . . 405</p><p align="left">10.8.1 One-Way ANOVA . . . . . . . . . . . . . . . . . . . . . . 407</p><p align="left">10.9 Power Analysis for Multivariate GLMs . . . . . . . . . . . . . . . 409</p><p align="left">10.9.1 Two Groups . . . . . . . . . . . . . . . . . . . . . . . . . . 409</p><p align="left">10.9.2 Repeated Measures Design . . . . . . . . . . . . . . . . . . 409</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">11 Two-level Hierarchical Linear Models 413</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413</p><p align="left">11.2 Two-level Hierarchical Linear Models . . . . . . . . . . . . . . . . 413</p><p align="left">11.3 Random Coefficient Model: One Population . . . . . . . . . . . . . 424</p><p align="left">11.4 Random Coefficient Model: Several Populations . . . . . . . . . . . 435</p><p align="left">11.5 Mixed Model Repeated Measures . . . . . . . . . . . . . . . . . . 440</p><p align="left">11.6 Mixed Model Repeated Measures with Changing Covariates . . . . 442</p><p align="left">11.7 Two-Level Hierarchical Linear Models . . . . . . . . . . . . . . . . 443</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">12 Incomplete Repeated Measurement Data 455</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">12.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455</p><p align="left">12.2 Missing Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . 456</p><p align="left">12.3 An FGLS Procedure . . . . . . . . . . . . . . . . . . . . . . . . . 457</p><p align="left">12.4 An ML Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . 460</p><p align="left">12.5 Imputations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 461</p><p align="left">12.5.1 EM Algorithm . . . . . . . . . . . . . . . . . . . . . . . . 462</p><p align="left">12.5.2 Multiple Imputation . . . . . . . . . . . . . . . . . . . . . 463</p><p align="left">12.6 Repeated Measures Analysis . . . . . . . . . . . . . . . . . . . . . 464</p><p align="left">12.7 Repeated Measures with Changing Covariates . . . . . . . . . . . . 464</p><p align="left">12.8 Random Coefficient Model . . . . . . . . . . . . . . . . . . . . . . 467</p><p>12.9 Growth Curve Analysis . . . . . . . . . . . . . . . . . . . . . . . . 471</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">13 Structural Equation Modeling 479</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">13.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479</p><p align="left">13.2 Model Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 481</p><p align="left">13.3 Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 489</p><p align="left">13.4 Model Fit in Practice . . . . . . . . . . . . . . . . . . . . . . . . . 494</p><p align="left">13.5 Model Modification . . . . . . . . . . . . . . . . . . . . . . . . . . 496</p><p align="left">13.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 498</p><p align="left">13.7 Path Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 499</p><p align="left">13.8 Confirmatory Factor Analysis . . . . . . . . . . . . . . . . . . . . . 503</p><p align="left">13.9 General SEM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 503</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p>References 511</p></font></font></font></font></font></font></font></font></p></font></font><font face="CMMI10" size="2"><font face="CMMI10" size="2">df </font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">for the </font></font><font face="CMMI10" size="2"><font face="CMMI10" size="2">F </font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">statistic in GLMMs . . . . . . . 118<p align="left">4.5 Maximum Likelihood Estimation and Fisher’s Information Matrix . 119</p><p align="left">4.6 WLSE for data Heteroscedasticity . . . . . . . . . . . . . . . . . . 121</p><p align="left">4.7 WLSE for Correlated Errors . . . . . . . . . . . . . . . . . . . . . 124</p><p align="left">4.8 FGLS for Categorical Data . . . . . . . . . . . . . . . . . . . . . . 127</p><p align="left">4.8.1 Overview of the Categorical Data Model . . . . . . . . . . 127</p><p align="left">4.8.2 Marginal Homogeneity . . . . . . . . . . . . . . . . . . . . 130</p><p align="left">4.8.3 Homogeneity of Proportions . . . . . . . . . . . . . . . . . 132</p><p align="left">4.8.4 Independence . . . . . . . . . . . . . . . . . . . . . . . . . 138</p><p align="left">4.8.5 Univariate Mixed Linear Model, Less than Full Rank Representation</p><p>. . . . . . . . . . . . . . . . . . . . . . . . . . . 141</p><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">5 Multivariate General Linear Models 143</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143</p><p align="left">5.2 Developing the Model . . . . . . . . . . . . . . . . . . . . . . . . 143</p><p align="left">5.3 Estimation Theory and Hypothesis Testing . . . . . . . . . . . . . . 145</p><p align="left">5.4 Multivariate Regression . . . . . . . . . . . . . . . . . . . . . . . . 152</p><p align="left">5.5 Classical and Normal Multivariate Linear Regression Models . . . . 153</p><p align="left">5.6 Jointly Multivariate Normal Regression Model . . . . . . . . . . . 163</p><p align="left">5.7 Multivariate Mixed Models and the Analysis of Repeated Measurements</p><p align="left">. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171</p><p align="left">5.8 Extended Linear Hypotheses . . . . . . . . . . . . . . . . . . . . . 176</p><p align="left">5.9 Multivariate Regression: Calibration and Prediction . . . . . . . . . 182</p><p align="left">5.9.1 Fixed <font face="CMMI10" size="2"><font face="CMMI10" size="2">X </font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">. . . . . . . . . . . . . . . . . . . . . . . . . . . . 182<p align="left">5.9.2 Random <font face="CMMI10" size="2"><font face="CMMI10" size="2">X </font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">. . . . . . . . . . . . . . . . . . . . . . . . . . 185<p align="left">5.9.3 Random <font face="CMMI10" size="2"><font face="CMMI10" size="2">X</font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">, Prediction . . . . . . . . . . . . . . . . . . . . 186<p align="left">5.9.4 Overview - Candidate Model . . . . . . . . . . . . . . . . . 186</p><p align="left">5.9.5 Prediction and Shrinkage . . . . . . . . . . . . . . . . . . . 187</p><p align="left">5.10 Multivariate Regression: Influential Observations . . . . . . . . . . 189</p><p align="left">5.10.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 191</p><p align="left">5.11 Nonorthogonal MANOVA designs . . . . . . . . . . . . . . . . . . 192</p><p align="left">5.11.1 Unweighted Analysis . . . . . . . . . . . . . . . . . . . . . 197</p><p align="left">5.11.2 Weighted Analysis . . . . . . . . . . . . . . . . . . . . . . 198</p><p align="left">5.12 MANCOVA Designs . . . . . . . . . . . . . . . . . . . . . . . . . 200</p><p align="left">5.12.1 Overall tests . . . . . . . . . . . . . . . . . . . . . . . . . 200</p><p align="left">5.12.2 Tests of Additional Information . . . . . . . . . . . . . . . 203</p><p align="left">5.12.3 Results and Interpretation . . . . . . . . . . . . . . . . . . 204</p><p align="left">5.13 Stepdown Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 206</p><p align="left">5.14 Repeated Measures Analysis . . . . . . . . . . . . . . . . . . . . . 208</p><p align="left">5.14.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 209</p><p align="left">5.15 Extended Linear Hypotheses . . . . . . . . . . . . . . . . . . . . . 216</p><p align="left">5.15.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 219</p></font></font></p></font></font></p></font></font></p></font></font><font face="CMMI10" size="2"><font face="CMMI10" size="2">X </font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">. . . . . . . . . . . . . . . . . . . . . . . . . . . . 182<p align="left">5.9.2 Random <font face="CMMI10" size="2"><font face="CMMI10" size="2">X </font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">. . . . . . . . . . . . . . . . . . . . . . . . . . 185<p align="left">5.9.3 Random <font face="CMMI10" size="2"><font face="CMMI10" size="2">X</font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">, Prediction . . . . . . . . . . . . . . . . . . . . 186<p align="left">5.9.4 Overview - Candidate Model . . . . . . . . . . . . . . . . . 186</p><p align="left">5.9.5 Prediction and Shrinkage . . . . . . . . . . . . . . . . . . . 187</p><p align="left">5.10 Multivariate Regression: Influential Observations . . . . . . . . . . 189</p><p align="left">5.10.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 191</p><p align="left">5.11 Nonorthogonal MANOVA designs . . . . . . . . . . . . . . . . . . 192</p><p align="left">5.11.1 Unweighted Analysis . . . . . . . . . . . . . . . . . . . . . 197</p><p align="left">5.11.2 Weighted Analysis . . . . . . . . . . . . . . . . . . . . . . 198</p><p align="left">5.12 MANCOVA Designs . . . . . . . . . . . . . . . . . . . . . . . . . 200</p><p align="left">5.12.1 Overall tests . . . . . . . . . . . . . . . . . . . . . . . . . 200</p><p align="left">5.12.2 Tests of Additional Information . . . . . . . . . . . . . . . 203</p><p align="left">5.12.3 Results and Interpretation . . . . . . . . . . . . . . . . . . 204</p><p align="left">5.13 Stepdown Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 206</p><p align="left">5.14 Repeated Measures Analysis . . . . . . . . . . . . . . . . . . . . . 208</p><p align="left">5.14.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 209</p><p align="left">5.15 Extended Linear Hypotheses . . . . . . . . . . . . . . . . . . . . . 216</p><p align="left">5.15.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 219</p></font></font></p></font></font></p></font></font><font face="CMMI10" size="2"><font face="CMMI10" size="2">X </font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">. . . . . . . . . . . . . . . . . . . . . . . . . . 185<p align="left">5.9.3 Random <font face="CMMI10" size="2"><font face="CMMI10" size="2">X</font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">, Prediction . . . . . . . . . . . . . . . . . . . . 186<p align="left">5.9.4 Overview - Candidate Model . . . . . . . . . . . . . . . . . 186</p><p align="left">5.9.5 Prediction and Shrinkage . . . . . . . . . . . . . . . . . . . 187</p><p align="left">5.10 Multivariate Regression: Influential Observations . . . . . . . . . . 189</p><p align="left">5.10.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 191</p><p align="left">5.11 Nonorthogonal MANOVA designs . . . . . . . . . . . . . . . . . . 192</p><p align="left">5.11.1 Unweighted Analysis . . . . . . . . . . . . . . . . . . . . . 197</p><p align="left">5.11.2 Weighted Analysis . . . . . . . . . . . . . . . . . . . . . . 198</p><p align="left">5.12 MANCOVA Designs . . . . . . . . . . . . . . . . . . . . . . . . . 200</p><p align="left">5.12.1 Overall tests . . . . . . . . . . . . . . . . . . . . . . . . . 200</p><p align="left">5.12.2 Tests of Additional Information . . . . . . . . . . . . . . . 203</p><p align="left">5.12.3 Results and Interpretation . . . . . . . . . . . . . . . . . . 204</p><p align="left">5.13 Stepdown Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 206</p><p align="left">5.14 Repeated Measures Analysis . . . . . . . . . . . . . . . . . . . . . 208</p><p align="left">5.14.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 209</p><p align="left">5.15 Extended Linear Hypotheses . . . . . . . . . . . . . . . . . . . . . 216</p><p align="left">5.15.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 219</p></font></font></p></font></font><font face="CMMI10" size="2"><font face="CMMI10" size="2">X</font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2">, Prediction . . . . . . . . . . . . . . . . . . . . 186<p align="left">5.9.4 Overview - Candidate Model . . . . . . . . . . . . . . . . . 186</p><p align="left">5.9.5 Prediction and Shrinkage . . . . . . . . . . . . . . . . . . . 187</p><p align="left">5.10 Multivariate Regression: Influential Observations . . . . . . . . . . 189</p><p align="left">5.10.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 191</p><p align="left">5.11 Nonorthogonal MANOVA designs . . . . . . . . . . . . . . . . . . 192</p><p align="left">5.11.1 Unweighted Analysis . . . . . . . . . . . . . . . . . . . . . 197</p><p align="left">5.11.2 Weighted Analysis . . . . . . . . . . . . . . . . . . . . . . 198</p><p align="left">5.12 MANCOVA Designs . . . . . . . . . . . . . . . . . . . . . . . . . 200</p><p align="left">5.12.1 Overall tests . . . . . . . . . . . . . . . . . . . . . . . . . 200</p><p align="left">5.12.2 Tests of Additional Information . . . . . . . . . . . . . . . 203</p><p align="left">5.12.3 Results and Interpretation . . . . . . . . . . . . . . . . . . 204</p><p align="left">5.13 Stepdown Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 206</p><p align="left">5.14 Repeated Measures Analysis . . . . . . . . . . . . . . . . . . . . . 208</p><p align="left">5.14.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 209</p><p align="left">5.15 Extended Linear Hypotheses . . . . . . . . . . . . . . . . . . . . . 216</p><p align="left">5.15.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 219</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">6 Doubly Multivariate Linear Model 223</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223</p><p align="left">6.2 Classical Model Development . . . . . . . . . . . . . . . . . . . . 223</p><p align="left">6.3 Responsewise Model Development . . . . . . . . . . . . . . . . . . 226</p><p align="left">6.4 The Multivariate Mixed Model . . . . . . . . . . . . . . . . . . . . 227</p><p align="left">6.5 Double Multivariate and Mixed Models . . . . . . . . . . . . . . . 231</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">7 The Restricted MGLM and the Growth Curve Model 243</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243</p><p align="left">7.2 The Restricted Multivariate General Linear Model . . . . . . . . . . 243</p><p align="left">7.3 The GMANOVA Model . . . . . . . . . . . . . . . . . . . . . . . 247</p><p align="left">7.4 Canonical Form of the GMANOVA Model . . . . . . . . . . . . . . 253</p><p align="left">7.5 Restricted Nonorthogonal Three-Factor Factorial MANOVA . . . . 259</p><p>7.5.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 269</p><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">7.6 Restricted Intraclass Covariance Design . . . . . . . . . . . . . . . 269</p><p align="left">7.6.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 275</p><p align="left">7.7 Growth Curve Analysis . . . . . . . . . . . . . . . . . . . . . . . . 279</p><p align="left">7.7.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 283</p><p align="left">7.8 Multiple Response Growth Curves . . . . . . . . . . . . . . . . . . 289</p><p align="left">7.8.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 290</p><p align="left">7.9 Single Growth Curve . . . . . . . . . . . . . . . . . . . . . . . . . 294</p><p align="left">7.9.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 294</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">8 The SUR Model and the Restricted GMANOVA model 297</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297</p><p align="left">8.2 The MANOVA-GMANOVA Model . . . . . . . . . . . . . . . . . 297</p><p align="left">8.3 Tests of Fit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303</p><p align="left">8.4 Sum of Profiles and CGMANOVA Models . . . . . . . . . . . . . . 305</p><p align="left">8.5 The SUR Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307</p><p align="left">8.6 The Restricted GMANOVA Model . . . . . . . . . . . . . . . . . . 314</p><p align="left">8.7 GMANOVA-SUR: One Population . . . . . . . . . . . . . . . . . . 317</p><p align="left">8.7.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 317</p><p align="left">8.8 GMANOVA-SUR: Several Populations . . . . . . . . . . . . . . . 319</p><p align="left">8.8.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 319</p><p align="left">8.9 SUR Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319</p><p align="left">8.9.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 323</p><p align="left">8.10 Two-Period Crossover Design with Changing Covariates . . . . . . 328</p><p align="left">8.10.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 329</p><p align="left">8.11 Repeated Measurements with Changing Covariates . . . . . . . . . 334</p><p align="left">8.11.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 335</p><p align="left">8.12 MANOVA-GMANOVA Model . . . . . . . . . . . . . . . . . . . . 337</p><p align="left">8.12.1 Results and interpretation . . . . . . . . . . . . . . . . . . 338</p><p align="left">8.13 CGMANOVA Model . . . . . . . . . . . . . . . . . . . . . . . . . 344</p><p align="left">8.13.1 Results and Interpretation . . . . . . . . . . . . . . . . . . 346</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">9 Simultaneous Inference Using Finite Intersection Tests 349</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349</p><p align="left">9.2 Finite Intersection Tests . . . . . . . . . . . . . . . . . . . . . . . . 349</p><p align="left">9.3 Finite Intersection Tests of Univariate Means . . . . . . . . . . . . 350</p><p align="left">9.4 Finite Intersection Tests for Linear Models . . . . . . . . . . . . . . 354</p><p align="left">9.5 A Comparisons of Some Tests of Univariate Means . . . . . . . . . 355</p><p align="left">9.5.1 Single-Step Methods . . . . . . . . . . . . . . . . . . . . . 355</p><p align="left">9.5.2 Stepdown Methods . . . . . . . . . . . . . . . . . . . . . . 357</p><p align="left">9.6 Analysis of Means Analysis . . . . . . . . . . . . . . . . . . . . . 358</p><p align="left">9.7 Simultaneous Test Procedures for Mean Vectors . . . . . . . . . . . 360</p><p align="left">9.8 Finite Intersection Test of Mean Vectors . . . . . . . . . . . . . . . 362</p><p align="left">9.9 Finite Intersection Test of Mean Vectors with Covariates . . . . . . 366</p><p align="left">9.10 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368</p><p>9.11 Univariate: One-way ANOVA . . . . . . . . . . . . . . . . . . . . 369</p><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">9.12 Multivariate: One-way MANOVA . . . . . . . . . . . . . . . . . . 372</p><p align="left">9.13 Multivariate: One-way MANCOVA . . . . . . . . . . . . . . . . . 379</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">10 Computing Power for Univariate and Multivariate GLM 381</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381</p><p align="left">10.2 Power for Univariate GLMs . . . . . . . . . . . . . . . . . . . . . 383</p><p align="left">10.3 Estimating Power, Sample Size, and Effect Size for the GLM . . . . 384</p><p align="left">10.3.1 Power and Sample Size . . . . . . . . . . . . . . . . . . . . 384</p><p align="left">10.3.2 Effect Size . . . . . . . . . . . . . . . . . . . . . . . . . . 385</p><p align="left">10.4 Power and Sample Size based upon Interval-Estimation . . . . . . . 388</p><p align="left">10.5 Calculating Power and Sample Size for Some Mixed Models . . . . 390</p><p align="left">10.5.1 Random One-Way ANOVA Design . . . . . . . . . . . . . 390</p><p align="left">10.5.2 Two Factor Mixed Nested ANOVA Design . . . . . . . . . 396</p><p align="left">10.6 Power for Multivariate GLMs . . . . . . . . . . . . . . . . . . . . 400</p><p align="left">10.7 Power and Effect Size Analysis for Univariate GLMs . . . . . . . . 401</p><p align="left">10.7.1 One-Way ANOVA . . . . . . . . . . . . . . . . . . . . . . 401</p><p align="left">10.7.2 Three-Way ANOVA . . . . . . . . . . . . . . . . . . . . . 403</p><p align="left">10.7.3 One-Way ANCOVA Design with two covariates . . . . . . 405</p><p align="left">10.8 Power and Sample Size based upon Interval-Estimation . . . . . . . 405</p><p align="left">10.8.1 One-Way ANOVA . . . . . . . . . . . . . . . . . . . . . . 407</p><p align="left">10.9 Power Analysis for Multivariate GLMs . . . . . . . . . . . . . . . 409</p><p align="left">10.9.1 Two Groups . . . . . . . . . . . . . . . . . . . . . . . . . . 409</p><p align="left">10.9.2 Repeated Measures Design . . . . . . . . . . . . . . . . . . 409</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">11 Two-level Hierarchical Linear Models 413</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413</p><p align="left">11.2 Two-level Hierarchical Linear Models . . . . . . . . . . . . . . . . 413</p><p align="left">11.3 Random Coefficient Model: One Population . . . . . . . . . . . . . 424</p><p align="left">11.4 Random Coefficient Model: Several Populations . . . . . . . . . . . 435</p><p align="left">11.5 Mixed Model Repeated Measures . . . . . . . . . . . . . . . . . . 440</p><p align="left">11.6 Mixed Model Repeated Measures with Changing Covariates . . . . 442</p><p align="left">11.7 Two-Level Hierarchical Linear Models . . . . . . . . . . . . . . . . 443</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">12 Incomplete Repeated Measurement Data 455</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">12.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455</p><p align="left">12.2 Missing Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . 456</p><p align="left">12.3 An FGLS Procedure . . . . . . . . . . . . . . . . . . . . . . . . . 457</p><p align="left">12.4 An ML Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . 460</p><p align="left">12.5 Imputations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 461</p><p align="left">12.5.1 EM Algorithm . . . . . . . . . . . . . . . . . . . . . . . . 462</p><p align="left">12.5.2 Multiple Imputation . . . . . . . . . . . . . . . . . . . . . 463</p><p align="left">12.6 Repeated Measures Analysis . . . . . . . . . . . . . . . . . . . . . 464</p><p align="left">12.7 Repeated Measures with Changing Covariates . . . . . . . . . . . . 464</p><p align="left">12.8 Random Coefficient Model . . . . . . . . . . . . . . . . . . . . . . 467</p><p>12.9 Growth Curve Analysis . . . . . . . . . . . . . . . . . . . . . . . . 471</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p align="left">13 Structural Equation Modeling 479</p></font></font><font face="NimbusRomNo9L-Regu" size="2"><font face="NimbusRomNo9L-Regu" size="2"><p align="left">13.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479</p><p align="left">13.2 Model Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 481</p><p align="left">13.3 Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 489</p><p align="left">13.4 Model Fit in Practice . . . . . . . . . . . . . . . . . . . . . . . . . 494</p><p align="left">13.5 Model Modification . . . . . . . . . . . . . . . . . . . . . . . . . . 496</p><p align="left">13.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 498</p><p align="left">13.7 Path Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 499</p><p align="left">13.8 Confirmatory Factor Analysis . . . . . . . . . . . . . . . . . . . . . 503</p><p align="left">13.9 General SEM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 503</p></font></font><font face="NimbusRomNo9L-Medi" size="2"><font face="NimbusRomNo9L-Medi" size="2"><p>References 511</p></font></font></font></font></font></font></font></font></font></font>
|