Statrep

Welcome to the StatRep Documentation.

The pdf documentation is the most complete; this page will get you started. See statrepmanual.pdf for full usage details and statrep.pdf for the implementation.

Installation

Complete installation instructions are included in the README and the user's manual. If you just want a quick look to see how it works, put the files in same directory as your LaTeX file (or use the included quickstart.tex file). As long as the *.sty files and the statrep_macros.sas file are in the current working directory, you can play with the package as much as you like. Then if you decide you want to use it for other documents, you can follow the complete installation instructions.

How it Works

When you compile your LaTeX document, the StatRep package parses your file and writes a SAS program from your code blocks and output tags. You run that program with SAS to create the actual output files. Then you compile your document again and the output files are included in the document.

Why Use It?

You can have complete confidence that the code you display in your document is the code that created the outputs, and you can recreate the outputs at any time from your LaTeX document and SAS. Your data and code reside in one LaTeX source file that you can share with others for reproducible research.

Getting Started

Assuming you have a fairly recent TeX distribution and SAS 9.2 or later installed, and that you have installed the StatRep package, consider the LaTeX document called quickstart.tex; it is part of the distribution so you already have a copy.

The main parts of the file are described in the following sections.

Load the class you want and the StatRep package:

\documentclass{article}
\usepackage{statrep}
\title{Example and Tutorial for the StatRep Package}
\author{Tim Arnold and Warren F. Kuhfeld\\SAS Institute Inc., Cary, NC}
\date{April 25, 2012}
\begin{document}
\maketitle
\section{Introduction}
This article provides an example and a
tutorial that show how to use the StatRep LaTeX package.

The purpose of the Datastep environment is to read in data. It produces no output. Its contents are written to the generated SAS program as-is.

\begin{Datastep}
data Wine;
   input WineType $ VisitLength @@;
   datalines;
white  80 white  98 white 115 white  89 white 103
white  91 white 119 white  31 white 109 white  95
white  71 white 105 white  66 white 141 white  79
white 113 white  69 white 120 white  87 red    93
red    87 red   106 red    76 red   121 red   143
red    81 red    97 red    74 red   107 red   112
red    67 red   107 red    72 red   116 red    99
red   104 red    91 red   132 red    78 red   107
red   101 red    92
;
\end{Datastep}

The purpose of the Sascode environment is to generate output. You use the store= option so that later in your document you can refer to output that is created in the Sascode environment. In this example, all output that is generated by the analysis is stored in the ODS document wineA .

With no other options or commands inside the Sascode environment, its contents are written to the generated SAS program as-is.

\begin{Sascode}[store=wineA]
ods graphics on;
proc anova data=Wine;
   class WineType;
   model VisitLength = WineType;
run;
ods graphics off;
\end{Sascode}

The Listing and Graphic tags specify the output to be displayed. The purpose of the Listing tag is to display tabular output and notes. The purpose of the Graphic tag is to display graphical output.

The Listing tag selects three output tables from the wineA ODS document: ClassLevels , NObs , and OverallANOVA . The Graphic tag selects the BoxPlot graph from the wineA ODS document.

The tags are parsed by the StatRep package and appropriate SAS code is written to the generated program to write the outputs to an external file. Once the files have been created, the tags include the outputs into the document.

\Listing[store=wineA,
   objects=ClassLevels NObs OverallANOVA,
   caption={Analysis of Variance for Visit Length}]{tsta}

\Graphic[store=wineA,
   objects=BoxPlot,
caption={Box Plots for Visit Length}]{tstb}

When you generate the SAS program by compiling your document with pdfLaTeX, the StatRep package does the following:

• The lines in the Datastep environment are passed unchanged to the program.
• The lines in the Sascode environment are parsed for line commands and passed to the program.
• Each Listing tag selects the specified notes and tables.
• Each Graphic tag selects the specified graphs.

When you execute the generated SAS program, the output that is specified in the Listing and Graphic tags is automatically captured. When you recompile your document, the Listing and Graphic tags insert the requested SAS results and page breaks are handled automatically.