Regression

SIMPLE LINEAR REGRESSION

ROUTINE DESCRIPTION
RLINE Fits a line to a set of data points using least squares.
RONE Analyzes a simple linear regression model.
RINCF Performs response control given a fitted simple linear regression model.
RINPF Performs inverse prediction given a fitted simple linear regression model.
 

MULTIVARIATE GENERAL LINEAR MODEL ANALYSIS

MODEL FITTING

ROUTINE DESCRIPTION
RLSE Fits a multiple linear regression model using least squares.
RCOV Fits a multivariate linear regression model given the variance-covariance matrix.
RGIVN Fits a multivariate linear regression model via fast Givens transformations.
RGLM Fits a multivariate general linear model.
RLEQU Fits a multivariate linear regression model with linear equality restrictions H B = G imposed on the regression parameters given results from routine RGIVN after IDO = 1 and IDO = 2 and prior to IDO = 3.
 

STATISTICAL INFERENCE AND DIAGNOSTICS

ROUTINE DESCRIPTION
RSTAT Computes statistics related to a regression fit given the coefficient estimates.
RCOVB Computes the estimated variance-covariance matrix of the estimated regression coefficients given the R matrix.
CESTI Constructs an equivalent completely testable multivariate general linear hypothesis H BU = G from a partially testable hypothesis HpBU = Gp.
RHPSS Computes the matrix of sums of squares and crossproducts for the multivariate general linear hypothesis H BU = G given the coefficient estimates and the R matrix.
RHPTE Performs tests for a multivariate general linear hypothesis H BU = G given the hypothesis sums of squares and crossproducts matrix SH and the error sums of squares and crossproducts matrix SE.
RLOFE Computes a lack of fit test based on exact replicates for a fitted regression model.
RLOFN Computes a lack of fit test based on near replicates for a fitted regression model.
RCASE Computes case statistics and diagnostics given data points, coefficient estimates and the R matrix for a fitted general linear model.
ROTIN Computes diagnostics for detection of outliers and influential data points given residuals and the R matrix for a fitted general linear model.
 

UTILITIES FOR CLASSIFICATION VARIABLES

ROUTINE DESCRIPTION
GCLAS Gets the unique values of each classification variable.
GRGLM Generates regressors for a general linear model.
 

VARIABLES SELECTION

ROUTINE DESCRIPTION
RBEST Selects the best multiple linear regression models.
RSTEP Builds multiple linear regression models using forward selection, backward selection or stepwise selection.
GSWEP Performs a generalized sweep of a row of a nonnegative definite matrix.
RSUBM Retrieves a symmetric submatrix from a symmetric matrix.
 

POLYNOMINAL REGRESSION AND SECOND-ORDER MODELS

POLYNOMINAL REGRESSION ANALYSIS

ROUTINE DESCRIPTION
RCURV Fits a polynomial curve using least squares.
RPOLY Analyzes a polynomial regression model.
 

SECOND-ORDER MODEL DESIGN

ROUTINE DESCRIPTION
RCOMP Generates an orthogonal central composite design.
 

UTILITY ROUTINES FOR POLYNOMIAL MODELS AND SECOND-ORDER MODELS

ROUTINE DESCRIPTION
RFORP Fits an orthogonal polynomial regression model.
RSTAP Computes summary statistics for a polynomial regression model given the fit based on orthogonal polynomials.
RCASP Computes case statistics for a polynomial regression model given the fit based on orthogonal polynomials.
OPOLY Generates orthogonal polynomials with respect to x-values and specified weights.
GCSCP Generates centered variables, squares, and crossproducts.
TCSCP Transforms coefficients from a second order response surface model generated from squares and crossproducts of centered variables to a model using uncentered variables.
 

NONLINEAR REGRESSION ANALYSIS

ROUTINE DESCRIPTION
RNLIN Fits a nonlinear regression model.
 

FITTING LINEAR MODELS BASED ON CRITERIA OTHER THAN LEAST SQUARES

ROUTINE DESCRIPTION
RLAV Fits a multiple linear regression model using the least absolute values criterion.
RLLP Fits a multiple linear regression model using the Lp norm criterion.
RLMV Fits a multiple linear regression model using the minimax criterion.
PLSR Performs partial least squares regression for one or more response variables and one or more predictor variables.