12 ✦ Chapter 1: What’s New in SAS/ETS 9.22 The following tables are now available through the OUTPUT statement: E1, E2, E3, and E8. The SIGMALIM option of the X11 statement enables you to specify the upper and lower sigma limits that are used to identify and decrease the weight of extreme irregular values in the internal seasonal adjustment computations. The TYPE option of the X11 statement controls which factors are removed from the original series to produce the seasonally adjusted series (table D11) and also the final trend cycle (table D12). The OUTSTAT= option of the X12 statement specifies the optional output data set that contains the summary statistics related to each seasonally adjusted series. The data set is sorted by the BY-group variables, if any, and by series names. The PERIODOGRAM option of the X12 statement enables you to specify that the PERI- ODOGRAM rather than the SPECTRUM of the series be plotted in the G tables and plots. The PLOTS= option of the X12 statement controls the plots that are produced through ODS Graphics. The SPECTRUMSERIES option of the X12 statement specifies the table name of the series that is used in the spectrum of the original series (table G0). The table names that can be specified are A1, A19, B1, or E1. The default is B1. The following tables are now available through the TABLES statement: E1, E2, and E3. The following tables are now available through ODS: “Model Description for ARIMA Model Identification”, “Model Description for ARIMA Model Estimation”, “Final Seasonal Filter Selection via Global MSR”, “Seasonal Filters by Period”, and “Final Trend Cycle Statistics”. The model description information was previously displayed in notes; an ODS table enables you to export the information to a data set. The seasonal filter and trend filter tables are new. Auxiliary variables have been added to ACF and PACF data sets that are available through ODS OUTPUT. The following variables have been added: _NAME_, Transform, Adjust, Regressors, Diff, and Sdiff. The purpose of the new variables is to help you identify the source of the data when multiple ACFs and PACFs are calculated. The following new feature is experimental: The AUXDATA= option of the X12 specifies an auxiliary input data set that can contain user-defined variables specified in the INPUT statement, the USERVAR= option of the RE- GRESSION statment, or the USERDEFINED statement. The AUXDATA= option is useful when user-defined regressors are used for multiple time series data sets or multiple BY groups. SAS/ETS Model Editor Application (Experimental) A new interactive application, the SAS/ETS Model Editor, enables you to define, fit, and simulate nonlinear statistical models using the MODEL procedure. The SAS/ETS Model Editor enables you Date Intervals, Formats, and Functions ✦ 13 to use the powerful features of PROC MODEL through a convenient and interactive graphical user interface. Date Intervals, Formats, and Functions The custom time intervals that are available in Base SAS software can be used in SAS/ETS procedures. Custom time intervals enable you to specify beginning and ending dates and seasonality for time intervals according to any definition. Such intervals can be used to define the following: fiscal intervals such as monthly intervals that begin on a day other than the first day of the month (for example, intervals that begin on the 10th day of each month) fiscal intervals such as monthly intervals that begin on different days for different months (for example, March of 2000 can begin on March 10, but April of 2000 can begin on April 12) business days, such as banking days that exclude holidays hourly intervals that omit hours that the business is closed 14 Chapter 2 Introduction Contents Overview of SAS/ETS Software . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Uses of SAS/ETS Software . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Contents of SAS/ETS Software . . . . . . . . . . . . . . . . . . . . . . . . 18 About This Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Chapter Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Typographical Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Where to Turn for More Information . . . . . . . . . . . . . . . . . . . . . . . . . 22 Accessing the SAS/ETS Sample Library . . . . . . . . . . . . . . . . . . . 22 Online Help System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 SAS Short Courses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 SAS Technical Support Services . . . . . . . . . . . . . . . . . . . . . . . . 23 Major Features of SAS/ETS Software . . . . . . . . . . . . . . . . . . . . . . . . 23 Discrete Choice and Qualitative and Limited Dependent Variable Analysis . 23 Regression with Autocorrelated and Heteroscedastic Errors . . . . . . . . . 25 Simultaneous Systems Linear Regression . . . . . . . . . . . . . . . . . . . 26 Linear Systems Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Polynomial Distributed Lag Regression . . . . . . . . . . . . . . . . . . . . 28 Nonlinear Systems Regression and Simulation . . . . . . . . . . . . . . . . 29 ARIMA (Box-Jenkins) and ARIMAX (Box-Tiao) Modeling and Forecasting . 31 Vector Time Series Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 32 State Space Modeling and Forecasting . . . . . . . . . . . . . . . . . . . . 34 Spectral Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Seasonal Adjustment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Structural Time Series Modeling and Forecasting . . . . . . . . . . . . . . . 36 Time Series Cross-Sectional Regression Analysis . . . . . . . . . . . . . . . . 37 Automatic Time Series Forecasting . . . . . . . . . . . . . . . . . . . . . . 38 Time Series Interpolation and Frequency Conversion . . . . . . . . . . . . . 39 Trend and Seasonal Analysis on Transaction Databases . . . . . . . . . . . . 41 Access to Financial and Economic Databases . . . . . . . . . . . . . . . . . 42 Spreadsheet Calculations and Financial Report Generation . . . . . . . . . . 44 Loan Analysis, Comparison, and Amortization . . . . . . . . . . . . . . . . 45 Time Series Forecasting System . . . . . . . . . . . . . . . . . . . . . . . . 46 Investment Analysis System . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 ODS Graphics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 16 ✦ Chapter 2: Introduction Related SAS Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Base SAS Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 SAS Forecast Studio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 SAS High-Performance Forecasting . . . . . . . . . . . . . . . . . . . . . . 52 SAS/GRAPH Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 SAS/STAT Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 SAS/IML Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 SAS/IML Stat Studio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 SAS/OR Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 SAS/QC Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 MLE for User-Defined Likelihood Functions . . . . . . . . . . . . . . . . . 56 JMP Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 SAS Enterprise Guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 SAS Add-In for Microsoft Office . . . . . . . . . . . . . . . . . . . . . . . 59 Enterprise Miner—Time Series nodes . . . . . . . . . . . . . . . . . . . . . 59 SAS Risk Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Overview of SAS/ETS Software SAS/ETS software, a component of the SAS System, provides SAS procedures for: econometric analysis time series analysis time series forecasting systems modeling and simulation discrete choice analysis analysis of qualitative and limited dependent variable models seasonal adjustment of time series data financial analysis and reporting access to economic and financial databases time series data management In addition to SAS procedures, SAS/ETS software also includes seamless access to economic and financial databases and interactive environments for time series forecasting and investment analysis. Uses of SAS/ETS Software ✦ 17 Uses of SAS/ETS Software SAS/ETS software provides tools for a wide variety of applications in business, government, and academia. Major uses of SAS/ETS procedures are economic analysis, forecasting, economic and financial modeling, time series analysis, financial reporting, and manipulation of time series data. The common theme relating the many applications of the software is time series data: SAS/ETS software is useful whenever it is necessary to analyze or predict processes that take place over time or to analyze models that involve simultaneous relationships. Although SAS/ETS software is most closely associated with business, finance and economics, time series data also arise in many other fields. SAS/ETS software is useful whenever time dependencies, simultaneous relationships, or dynamic processes complicate data analysis. For example, an environ- mental quality study might use SAS/ETS software’s time series analysis tools to analyze pollution emissions data. A pharmacokinetic study might use SAS/ETS software’s features for nonlinear systems to model the dynamics of drug metabolism in different tissues. The diversity of problems for which econometrics and time series analysis tools are needed is reflected in the applications reported by SAS users. The following listed items are some applications of SAS/ETS software presented by SAS users at past annual conferences of the SAS Users Group International (SUGI). forecasting college enrollment (Calise and Earley 1997) fitting a pharmacokinetic model (Morelock et al. 1995) testing interaction effect in reducing sudden infant death syndrome (Fleming, Gibson, and Fleming 1996) forecasting operational indices to measure productivity changes (McCarty 1994) spectral decomposition and reconstruction of nuclear plant signals (Hoyer and Gross 1993) estimating parameters for the constant-elasticity-of-substitution translog model (Hisnanick 1993) applying econometric analysis for mass appraisal of real property (Amal and Weselowski 1993) forecasting telephone usage data (Fishetti, Heathcote, and Perry 1993) forecasting demand and utilization of inpatient hospital services (Hisnanick 1992) using conditional demand estimation to determine electricity demand (Keshani and Taylor 1992) estimating tree biomass for measurement of forestry yields (Parresol and Thomas 1991) evaluating the theory of input separability in the production function of U.S. manufacturing (Hisnanick 1991) 18 ✦ Chapter 2: Introduction forecasting dairy milk yields and composition (Benseman 1990) predicting the gloss of coated aluminum products subject to weathering (Khan 1990) learning curve analysis for predicting manufacturing costs of aircraft (Le Bouton 1989) analyzing Dow Jones stock index trends (Early, Sweeney, and Zekavat 1989) analyzing the usefulness of the composite index of leading economic indicators for forecasting the economy (Lin and Myers 1988) Contents of SAS/ETS Software Procedures SAS/ETS software includes the following SAS procedures: ARIMA ARIMA (Box-Jenkins) and ARIMAX (Box-Tiao) modeling and forecasting AUTOREG regression analysis with autocorrelated or heteroscedastic errors and ARCH and GARCH modeling COMPUTAB spreadsheet calculations and financial report generation COUNTREG regression modeling for dependent variables that represent counts DATASOURCE access to financial and economic databases ENTROPY maximum entropy-based regression ESM forecasting by using exponential smoothing models with optimized smoothing weights EXPAND time series interpolation, frequency conversion, and transformation of time series FORECAST automatic forecasting LOAN loan analysis and comparison MDC multinomial discrete choice analysis MODEL nonlinear simultaneous equations regression and nonlinear systems modeling and simulation PANEL panel data models PDLREG polynomial distributed lag regression QLIM qualitative and limited dependent variable analysis SIMILARITY similarity analysis of time series data for time series data mining SIMLIN linear systems simulation SPECTRA spectral and cross-spectral analysis STATESPACE state space modeling and automated forecasting of multivariate time series SYSLIN linear simultaneous equations models Contents of SAS/ETS Software ✦ 19 TIMESERIES analysis of time-stamped transactional data TSCSREG time series cross-sectional regression analysis UCM unobserved components analysis of time series VARMAX vector autoregressive and moving-average modeling and forecasting X11 seasonal adjustment (Census X-11 and X-11 ARIMA) X12 seasonal adjustment (Census X-12 ARIMA) Macros SAS/ETS software includes the following SAS macros: %AR generates statements to define autoregressive error models for the MODEL proce- dure %BOXCOXAR investigates Box-Cox transformations useful for modeling and forecasting a time series %DFPVALUE computes probabilities for Dickey-Fuller test statistics %DFTEST performs Dickey-Fuller tests for unit roots in a time series process %LOGTEST tests to determine whether a log transformation is appropriate for modeling and forecasting a time series %MA generates statements to define moving-average error models for the MODEL procedure %PDL generates statements to define polynomial distributed lag models for the MODEL procedure These macros are part of the SAS AUTOCALL facility and are automatically available for use in your SAS program. Refer to SAS Macro Language: Reference for information about the SAS macro facility. Access Interfaces to Economic and Financial Databases In addition to PROC DATASOURCE, these SAS/ETS access interfaces provide seamless access to financial and economic databases: SASECRSP LIBNAME engine for accessing time series and event data residing in CRSPAc- cess database. SASEFAME LIBNAME engine for accessing time or case series data residing in a FAME database. SASEHAVR LIBNAME engine for accessing time series residing in a HAVER ANALYTICS Data Link Express (DLX) database. 20 ✦ Chapter 2: Introduction The Time Series Forecasting System SAS/ETS software includes an interactive forecasting system, described in Part IV. This graphical user interface to SAS/ETS forecasting features was developed with SAS/AF software and uses PROC ARIMA and other internal routines to perform time series forecasting. The Time Series Forecasting System makes it easy to forecast time series and provides many features for graphical data exploration and graphical comparisons of forecasting models and forecasts. (You must have SAS/GRAPH ® installed to use the graphical features of the system.) The Investment Analysis System The Investment Analysis System, described in Part V, is an interactive environment for analyzing the time-value of money in a variety of investments. Various analyses are provided to help analyze the value of investment alternatives: time value, periodic equivalent, internal rate of return, benefit-cost ratio, and break-even analysis. About This Book This book is a user’s guide to SAS/ETS software. Since SAS/ETS software is a part of the SAS System, this book assumes that you are familiar with Base SAS software and have the books SAS Language Reference: Dictionary and Base SAS Procedures Guide available for reference. It also assumes that you are familiar with SAS data sets, the SAS DATA step, and with basic SAS procedures such as PROC PRINT and PROC SORT. Chapter 3, “Working with Time Series Data,” in this book summarizes the aspects of Base SAS software that are most relevant to the use of SAS/ETS software. Chapter Organization Following a brief What’s New, this book is divided into five major parts. Part I contains general information to aid you in working with SAS/ETS Software. Part II explains the SAS procedures of SAS/ETS software. Part III describes the available data access interfaces for economic and financial databases. Part IV is the reference for the Time Series Forecasting System, an interactive forecasting menu system that uses PROC ARIMA and other routines to perform time series forecasting. Finally, Part V is the reference for the Investment Analysis System. The new features added to SAS/ETS software since the publication of SAS/ETS Software: Changes and Enhancements for Release 8.2 are summarized in Chapter 1, “What’s New in SAS/ETS 9.22.” If you have used SAS/ETS software in the past, you may want to skim this chapter to see what’s new. Part I contains the following chapters. Chapter 2, the current chapter, provides an overview of SAS/ETS software and summarizes related SAS publications, products, and services. Typographical Conventions ✦ 21 Chapter 3, “Working with Time Series Data,” discusses the use of SAS data management and programming features for time series data. Chapter 4, “Date Intervals, Formats, and Functions,” summarizes the time intervals, date and datetime informats, date and datetime formats, and date and datetime functions available in the SAS System. Chapter 5, “SAS Macros and Functions,” documents SAS macros and DATA step financial functions provided with SAS/ETS software. The macros use SAS/ETS procedures to perform Dickey-Fuller tests, test for the need for log transformations, or select optimal Box-Cox transformation parameters for time series data. Chapter 6, “Nonlinear Optimization Methods,” documents the NonLinear Optimization subsystem used by some ETS procedures to perform nonlinear optimization tasks. Part II contains chapters that explain the SAS procedures that make up SAS/ETS software. These chapters appear in alphabetical order by procedure name. Part III contains chapters that document the ETS access interfaces to economic and financial databases. Each of the chapters that document the SAS/ETS procedures (Part II) and the SAS/ETS access interfaces (Part III) is organized as follows: 1. The “Overview” section gives a brief description of the procedure. 2. The “Getting Started” section provides a tutorial introduction on how to use the procedure. 3. The “Syntax” section is a reference to the SAS statements and options that control the procedure. 4. The “Details” section discusses various technical details. 5. The “Examples” section contains examples of the use of the procedure. 6. The “References” section contains technical references on methodology. Part IV contains the chapters that document the features of the Time Series Forecasting System. Part V contains chapters that document the features of the Investment Analysis System. Typographical Conventions This book uses several type styles for presenting information. The following list explains the meaning of the typographical conventions used in this book: roman is the standard type style used for most text. UPPERCASE ROMAN is used for SAS statements, options, and other SAS language elements when they appear in the text. However, you can enter these elements in . and Weselowski 199 3) forecasting telephone usage data (Fishetti, Heathcote, and Perry 199 3) forecasting demand and utilization of inpatient hospital services (Hisnanick 199 2) using conditional. reconstruction of nuclear plant signals (Hoyer and Gross 199 3) estimating parameters for the constant-elasticity-of-substitution translog model (Hisnanick 199 3) applying econometric analysis for mass appraisal. 199 5) testing interaction effect in reducing sudden infant death syndrome (Fleming, Gibson, and Fleming 199 6) forecasting operational indices to measure productivity changes (McCarty 199 4)