Fitting Linear Models • Standard Least Squares Report and Options
Previous
 • 
Next
Standard Least Squares Report and Options
Analyze Common Classes of Models
The Standard Least Squares personality within the Fit Model platform fits a wide spectrum of standard models. These models include regression, analysis of variance, analysis of covariance, and mixed models, as well as the models typically used to analyze designed experiments. Use the Standard Least Squares personality to construct linear models for continuous-response data using least squares or, in the case of random effects, restricted maximum likelihood (REML).
Analytic results are supported by compelling dynamic visualization tools such as profilers, contour plots, and surface plots (see the book Profilers). These visual displays stimulate, complement, and support your understanding of the model. They enable you to optimize several responses simultaneously and to explore the effect of noise.
Figure 2.1 Examples of Standard Least Squares Plots
Contents 
Example Using Standard Least Squares
Launch the Standard Least Squares Personality
Fit Model Launch Window
Standard Least Squares Options in the Fit Model Launch Window
Validation
Missing Values
Fit Least Squares Report
Single versus Multiple Responses
Report Structure Related to Emphasis
Special Reports
Least Squares Fit Options
Fit Group Options
Response Options
Regression Reports
Summary of Fit
Analysis of Variance
Parameter Estimates
Effect Tests
Effect Details
Lack of Fit
Estimates
Show Prediction Expression
Sorted Estimates
Expanded Estimates
Indicator Parameterization Estimates
Sequential Tests
Custom Test
Multiple Comparisons
Compare Slopes
Joint Factor Tests
Inverse Prediction
Cox Mixtures
Parameter Power
Correlation of Estimates
Coding for Nominal Effects
Effect Screening
Scaled Estimates and the Coding of Continuous Terms
Plot Options
Normal Plot Report
Bayes Plot Report
Pareto Plot Report
Factor Profiling
Profiler
Interaction Plots
Contour Profiler
Mixture Profiler
Cube Plots
Box Cox Y Transformation
Surface Profiler
Row Diagnostics
Leverage Plots
Press
Save Columns
Prediction Formula
Effect Summary Report
Mixed and Random Effect Model Reports and Options
Mixed Models and Random Effect Models
Restricted Maximum Likelihood (REML) Method
EMS (Traditional) Model Fit Reports
Models with Linear Dependencies among Model Terms
Singularity Details
Parameter Estimates Report
Effect Tests Report
Examples
Statistical Details for the Standard Least Squares Personality
Emphasis Rules
Details of Custom Test Example
Correlation of Estimates
Leverage Plot Details
The Kackar-Harville Correction
Power Analysis

Previous
 • 
Next
Help created on 7/12/2018