Language Reference

Overview

This chapter describes all operators, statements, functions, and subroutines that can be used in SAS/IML software. All necessary details, such as arguments and operands, are included.

This chapter is divided into two sections. The first section contains operator descriptions. They are in alphabetic order according to the name of the operator. The second section contains descriptions of statements, functions, and subroutines also arranged alphabetically by name.

The following tables list all statements, functions, and subroutines available in SAS/IML software grouped by functionality.

Scalar Functions

ABS Function	takes the absolute value
EXP Function	calculates the exponential
INT Function	truncates a value
LOG Function	takes the natural logarithm
MOD Function	computes the modulo (remainder)
NORMAL Function	generates a pseudo-random normal deviate
SQRT Function	calculates the square root
UNIFORM Function	generates pseudo-random uniform deviates

Reduction Functions

MAX Function	finds the maximum value of a matrix
MIN Function	finds the smallest element of a matrix
SSQ Function	calculates the sum of squares of all elements
SUM Function	sums all elements

Matrix Inquiry Functions

ALL Function	checks for all nonzero elements
ANY Function	checks for any nonzero elements
CHOOSE Function	conditionally chooses and changes elements
LOC Function	finds nonzero elements of a matrix
NCOL Function	finds the number of columns of a matrix
NLENG Function	finds the size of an element
NROW Function	finds the number of rows of a matrix
TYPE Function	determines the type of a matrix

Matrix Sorting and BY-Group Processing Calls

SORT Call	sorts a matrix by specified columns
SORTNDX Call	creates a sorted index for a matrix
UNIQUEBY Function	finds locations of unique BY groups in a sorted or indexed matrix

Matrix Reshaping Functions

BLOCK Function	forms block-diagonal matrices
BTRAN Function	computes block transpose
DIAG Function	creates a diagonal matrix
DO Function	produces an arithmetic series
I Function	creates an identity matrix
INSERT Function	inserts one matrix inside another
J Function	creates a matrix of identical values
REMOVE Function	discards elements from a matrix
REPEAT Function	creates a new matrix of repeated values
SHAPE Function	reshapes and repeats values
SQRSYM Function	converts a symmetric matrix to a square matrix
SYMSQR Function	converts a square matrix to a symmetric matrix
T Function	transposes a matrix
VECDIAG Function	creates a vector from a diagonal

Character Functionality

BYTE Function	translates numbers to ordinal characters
CHANGE Call	replaces text
CHAR Function	produces a character representation of a matrix
CONCAT Function	Concatenates elementwise strings
CSHAPE Function	reshapes and repeats character values
LENGTH Call	finds the lengths of character matrix elements
NAME Function	lists the names of arguments
NUM Function	produces a numeric representation of a character matrix
ROWCAT Function	concatenates rows without using blank compression
ROWCATC Function	concatenates rows by using blank compression
SUBSTR Function	takes substrings of matrix elements

Random Number Generation Functionality

RANDGEN Call	generates random numbers from specified distributions
RANDSEED Call	initializes seed for subsequent RANGEN calls

Statistical Functionality

BRANKS Function	computes bivariate ranks
CUSUM Function	calculates cumulative sums
DESIGN Function	creates a design matrix
DESIGNF Function	creates a full-rank design matrix
GEOMEAN Function	calculates geometric means
HADAMARD Function	creates a Hadamard matrix
HARMEAN Function	calculates harmonic means
IPF Call	performs an iterative proportional fit of a contingency table
LAV Call	performs linear least absolute value regression by solving the norm minimization problem
LMS Call	performs robust regression
LTS Call	performs robust regression
MAD Function	finds the univariate (scaled) median absolute deviation
MARG Call	evaluates marginal totals in a multiway contingency table
MAXQFORM Call	computes the subsets of a matrix system that maximize the quadratic form
MCD Call	finds the minimum covariance determinant estimator
MVE Call	finds the minimum volume ellipsoid estimator
OPSCAL Function	rescales qualitative data to be a least squares fit to qualitative data
RANK Function	ranks elements of a matrix
RANKTIE Function	ranks matrix elements by using tie-averaging
SEQSCALE Call	perform discrete sequential tests
SEQSHIFT Call	perform discrete sequential tests
SEQTESTS Calls	perform discrete sequential tests
SWEEP Function	sweeps a matrix

Time Series Functionality

ARMACOV Call	computes an autocovariance sequence for an ARMA model
ARMALIK Call	computes the log likelihood and residuals for an ARMA model
ARMASIM Function	simulates an ARMA series
CONVEXIT Function	calculates convexity of a noncontingent cash flow
COVLAG Function	computes autocovariance estimates for a vector time series
DURATION Function	calculates modified duration of a noncontingent cash flow
FARMACOV Call	computes the autocovariance function for an ARFIMA() process
FARMAFIT Call	estimate the parameters of an ARFIMA() model
FARMALIK Call	computes the log-likelihood function of an ARFIMA() model
FARMASIM Call	generates an ARFIMA() process
FDIF Call	obtain a fractionally differenced process
FORWARD Function	calculates forward rates
KALCVF Call	computes the one-step prediction and the filtered estimate , as well as their covariance matrices. The call uses forward recursions, and you can also use it to obtain -step estimates.
KALCVS Call	uses backward recursions to compute the smoothed estimate and its covariance matrix, , where is the number of observations in the complete data set.
KALDFF Call	computes the one-step forecast of state vectors in an SSM by using the diffuse Kalman filter. The call estimates the conditional expectation of , and it also estimates the initial random vector, , and its covariance matrix.
KALDFS Call	computes the smoothed state vector and its mean square error matrix from the one-step forecast and mean square error matrix computed by KALDFF.
PV Function	calculates present value
RATES Function	converts interest rates from one base to another
SPOT Function	calculates spot rates
TSBAYSEA Call	performs Bayesian seasonal adjustment modeling
TSDECOMP Call	analyzes nonstationary time series by using smoothness priors modeling
TSMLOCAR Call	analyzes nonstationary or locally stationary time series by using the minimum AIC procedure
TSMLOMAR Call	analyzes nonstationary or locally stationary multivariate time series by using the minimum AIC procedure
TSMULMAR Call	estimates VAR processes by using the minimum AIC procedure
TSPEARS Call	analyzes periodic AR models with the minimum AIC procedure
TSPRED Call	provides predicted values of univariate and multivariate ARMA processes when the ARMA coefficients are input
TSROOT Call	calculates AR and MA coefficients from the characteristic roots of the model or calculates the characteristic roots of the model from the AR and MA coefficients
TSTVCAR Call	analyzes time series that are nonstationary in the covariance function
TSUNIMAR Call	determines the order of an AR process with the minimum AIC procedure and estimates the AR coefficients
VARMACOV Call	computes the theoretical cross-covariance matrices for a stationary VARMA() model
VARMALIK Call	computes the log-likelihood function for a VARMA() model
VARMASIM Call	generates VARMA(,) time series
VNORMAL Call	generates multivariate normal random series
VTSROOT Call	computes the characteristic roots for a VARMA() model
YIELD Function	calculates yield-to-maturity of a cash-flow stream

Numerical Analysis Functionality

BSPLINE Function	computes B-spline basis
FFT Function	performs the finite Fourier transform
IFFT Function	computes the inverse finite Fourier transform
JROOT Function	computes the first nonzero roots of a Bessel function of the first kind and the derivative of the Bessel function at each root
ODE Call	performs numerical integration of vector differential equations of the form
ORPOL Function	generates orthogonal polynomials on a discrete set of points
ORTVEC Call	provides columnwise orthogonalization by the Gram-Schmidt process and stepwise QR decomposition by the Gram-Schmidt process
POLYROOT Function	finds zeros of a real polynomial
PRODUCT Function	multiplies matrices of polynomials
QUAD Call	performs numerical integration of scalar functions in one dimension over infinite, connected semi-infinite, and connected finite intervals
RATIO Function	divides matrix polynomials
SPLINE Call	fits a cubic spline to points
SPLINEC Call	fits a cubic spline to points
SPLINEV Function	evaluates a cubic spline at new data points
TPSPLINE Call	computes thin-plate smoothing splines
TPSPLNEV Call	evaluates the thin-plate smoothing spline at new data points

Linear Algebra Functionality

APPCORT CALL	completes orthogonal decomposition
COMPORT Call	completes orthogonal decomposition by Householder transformations
CVEXHULL Function	finds a convex hull of a set of planar points
DET Function	computes the determinant of a square matrix
ECHELON Function	reduces a matrix to row-echelon normal form
EIGEN Call	computes eigenvalues and eigenvectors
EIGVAL Function	computes eigenvalues
EIGVEC Function	computes eigenvectors
GENEIG Call	computes eigenvalues and eigenvectors of a generalized eigenproblem
GINV Function	computes the generalized inverse
GSORTH Call	computes the Gram-Schmidt orthonormalization
HALF Function	computes Cholesky decomposition
HANKEL Function	generates a Hankel matrix
HDIR Function	performs a horizontal direct product
HERMITE Function	reduces a matrix to Hermite normal form
HOMOGEN Function	solves homogeneous linear systems
INV Function	produces the inverse
INVUPDT Function	updates a matrix inverse
ITSOLVER Call	solves a sparse general linear system by iteration
LUPDT Call	provides updating and downdating for rank deficient linear least squares solutions, complete orthogonal factorization, and Moore-Penrose inverses
QR Call	produces the QR decomposition of a matrix by Householder transformations
RDODT Call	downdates and updates QR and Cholesky decompositions
ROOT Function	performs the Cholesky decomposition of a matrix
RUPDT Call	updates QR and Cholesky decompositions
RZLIND Call	updates QR and Cholesky decompositions
SOLVE Function	solves a system of linear equations
SOLVELIN Call	solves a sparse symmetric system of linear equations by direct decomposition
SVD Call	computes the singular value decomposition
TEIGEN Call	computes the eigenvalues and eigenvectors of square matrices
TEIGVAL Function	computes eigenvalues of square matrices
TEIGVEC Function	computes eigenvectors of square matrices
TOEPLITZ Function	generates a Toeplitz or block-Toeplitz matrix
TRACE Function	sums diagonal elements
TRISOLV Function	solves linear systems with triangular matrices
XMULT Function	performs accurate matrix multiplication

Optimization Subroutines

LCP Call	solves the linear complementarity problem
LP Call	solves the linear programming problem
NLPCG Call	performs nonlinear optimization by conjugate gradient method
NLPDD Call	performs nonlinear optimization by double-dogleg method
NLPFDD Call	approximates derivatives by finite-differences method
NLPFEA Call	computes feasible points subject to constraints
NLPHQN Call	calculates hybrid quasi-Newton least squares
NLPLM Call	calculates Levenberg-Marquardt least squares
NLPNMS Call	performs nonlinear optimization by Nelder-Mead simplex method
NLPNRA Call	performs nonlinear optimization by Newton-Raphson method
NLPNRR Call	performs nonlinear optimization by Newton-Raphson ridge method
NLPQN Call	performs nonlinear optimization by quasi-Newton method
NLPQUA Call	performs nonlinear optimization by quadratic method
NLPTR Call	performs nonlinear optimization by trust-region method
Nonlinear Optimization and Related Subroutines	lists of all nonlinear optimization and related subroutines in IML

Set Functions

SETDIF Function	compares elements of two matrices
UNION Function	performs unions of sets
UNIQUE Function	sorts and removes duplicates
XSECT Function	intersects sets

Control Statements

ABORT Statement	ends IML
APPLY Function	applies an IML module
CALL Statement	calls a subroutine or function
DO and END Statements	groups statements as a unit
DO, Iterative Statement	iteratively executes a DO group
DO and UNTIL Statement and Clause	conditionally executes statements iteratively
DO and WHILE Statement and Clause	conditionally executes statements iteratively
END Statement	ends a DO loop or DO statement
EXECUTE Call	executes SAS statements immediately
FINISH Statement	denotes the end of a module
FORCE Statement	(see the description of the SAVE statement)
FREE Statement	frees matrix storage space
GOTO Statement	jumps to a new statement
IF-THEN/ELSE Statement	conditionally executes statement
LINK Statement	jumps to another statement
MATTRIB Statement	associates printing attributes with matrices
PAUSE Statement	interrupts module execution
PRINT Statement	prints matrix values
PURGE Statement	removes observations marked for deletion and renumbers records
PUSH Call	pushes SAS statements into the command input stream
QUEUE Call	queues SAS statements into the command input stream
QUIT Statement	exits from IML
REMOVE Statement	removes matrices from storage
RESET Statement	sets processing options
RESUME Statement	resumes execution
RETURN Statement	returns to caller
RUN Statement	executes statements in a module
SHOW Statement	prints system information
SOUND Call	produces a tone
START/FINISH Statements	define a module
STOP Statement	stops execution of statements
STORAGE Function	lists names of matrices and modules in storage
STORE Statement	stores matrices and modules in library storage
VALSET Call	performs indirect assignment
VALUE Function	assigns values by indirect reference

Datas Set and File Functionality

APPEND Statement	adds observations to SAS data set
CLOSE Statement	closes a SAS data set
CLOSEFILE Statement	closes a file
CONTENTS Function	returns the variables in a SAS data set
CREATE Statement	creates a new SAS data set
DATASETS Function	obtains the names of SAS data sets
DELETE Call	deletes a SAS data set
DELETE Statement	marks observations for deletion
DO DATA Statement	repeats a loop until an end of file occurs
EDIT Statement	opens a SAS data set for editing
FILE Statement	opens or points to an external file
FIND Statement	finds observations
INDEX Statement	indexes a variable in a SAS data set
INFILE Statement	opens a file for input
INPUT Statement	inputs data
LIST Statement	displays observations of a data set
LOAD Statement	loads modules and matrices from library storage
PUT Statement	writes data to an external file
READ Statement	reads observations from a data set
RENAME Call	renames a SAS data set
REPLACE Statement	replaces values in observations and updates observations
SAVE Statement	saves data
SETIN Statement	makes a data set current for input
SETOUT Statement	makes a data set current for output
SORT Statement	sorts a SAS data set
SUMMARY Statement	computes summary statistics for SAS data sets
USE Statement	opens a SAS data set for reading

Graphics and Window Functions

DISPLAY Statement	displays fields in a display window
GBLKVP Call	defines a blanking viewport
GBLKVPD Call	deletes the blanking viewport
GCLOSE Call	closes the graphics segment
GDELETE Call	deletes a graphics segment
GDRAW Call	draws a polyline
GDRAWL Call	draws individual lines
GGRID Call	draws a grid
GINCLUDE Call	includes a graphics segment
GOPEN Call	opens a graphics segment
GPIE Call	draws pie slices
GPIEXY Call	converts from polar to world coordinates
GPOINT Call	plots points
GPOLY Call	draws and fills a polygon
GPORT Call	defines a viewport
GPORTPOP Call	pops the viewport
GPORTSTK Call	stacks the viewport
GSCALE Call	calculates round numbers for labeling axes
GSCRIPT Call	writes multiple text strings with special fonts
GSET Call	sets attributes for a graphics segment
GSHOW Call	shows a graph
GSTART Call	initializes the graphics system
GSTOP Call	deactivates the graphics system
GSTRLEN Call	finds the string length
GTEXT and GVTEXT Calls	places text horizontally or vertically on a graph
GWINDOW Call	defines the data window
GXAXIS and GYAXIS Calls	draws a horizontal or vertical axis
PGRAF Call	produces scatter plots
ODSGRAPH Call	renders a graph by using ODS Statistical Graphics
WINDOW Statement	opens a display window

Wavelet Analysis Calls

WAVFT Call	computes a specified wavelet transform of one dimensional data
WAVGET Call	returns requested information encapsulated in a wavelet transform
WAVIFT Call	inverts a wavelet transform after applying specified thresholding to the detail coefficients
WAVPRINT Call	displays requested information encapsulated in a wavelet transform
WAVTHRSH Call	applies specified thresholding to the detail coefficients of a wavelet transform

Genetic Algorithm Functionality

GAEND Call	terminates genetic algorithm and frees memory resources
GAGETMEM Call	gets requested members and objective values from current solution population
GAGETVAL Call	gets objective function values for requested member of current solution population
GAINIT Call	initializes the initial solution population
GAREEVAL Call	reevaluates the objective function for all solutions in current population
GASETCRO Call	specifies a current crossover operator
GASETMUT Call	specifies a current mutation operator
GASETOBJ Call	specifies a current objective function
GASETSEL Call	specifies a current selection parameters
GASETUP Function	sets up a specific genetic algorithm optimization problem

Calling External Modules

MODULEI Call	calls an external routine without any return code
MODULEIC Call	calls an external routine that returns a character
MODULEIC Call	calls an external routine that returns a numeric value