home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!paladin.american.edu!auvm!UNC.BITNET!UNCSM1
- Message-ID: <SAS-L%92072015461152@VTVM2.BITNET>
- Newsgroups: bit.listserv.sas-l
- Date: Mon, 20 Jul 1992 15:45:00 EST
- Reply-To: Sally Muller <UNCSM1@UNC.BITNET>
- Sender: "SAS(r) Discussion" <SAS-L@UGA.BITNET>
- From: Sally Muller <UNCSM1@UNC.BITNET>
- Subject: TSCSREG available for PC SAS 6.04
- Lines: 84
-
- ----------------------------------------------------------------------
- SUMMARY: TSCSREG available for PC SAS 6.04
- REL/PLTF: 6.04 PC SAS
- E-ADDR: uncsm1@unc.bitnet
- NAME: Sally Muller
- PH/ADDR: 919-962-6501 OIT CB#3455, UNC, Chapel Hill NC 27514
- ----------------------------------------------------------------------
- Hey SAS-Lers!
- Looks like Paul OldenKamp may get his wish....long live PC/SAS!
- If you want it all you have to do is contact the Statistics group
- in SI Technical Support at 919-677-8008. Thanks to Dave Delong and
- Eddie Routen at SAS for making this available....
-
-
- ANNOUNCING THE RELEASE OF EXPERIMENTAL ENHANCEMENTS TO THE
- PC SAS/ETS TSCSREG PROCEDURE (6.04)
-
- Please note that you must be licensed for SAS/ETS for the TSCSREG
- procedure to execute. As with all experimental modules this
- procedure should be used with caution. Although it has been
- tested as part of the development process, it has not gone
- through the SAS Institute Quality Assurance testing process.
-
- The TSCSREG (Time Series Cross-Section Regression) procedure
- analyzes a class of linear econometric models that commonly arise
- when combining time series and cross-sectional data. The TSCSREG
- procedure deals with panel data on each of several cross
- sectional units. This procedure allows the user to estimate the
- regression parameters of the specified model under one or more of
- the following three common error structures:
-
- -- a variance components (or error components) model using the
- Fuller and Battese method of estimation (this is the default);
-
- -- a first-order autoregressive model with contemporaneous
- correlation using the Parks method of estimation;
-
- -- a mixed variance-component moving-average error process using
- the DaSilva method of estimation.
-
- Using PROC TSCSREG Statements
-
- The input data set used by the TSCSREG procedure must be sorted
- by cross sections and by time within each cross section. The
- variables that identify the cross section and the time period
- for each observation must be listed on an ID statement.
- Alternatively, one could use the TS= and CS= options in the
- PROC TSCSREG statement to specify the number of time series (TS)
- observations in each cross section and the number of cross
- sections (CS) in the data set.
-
- The regression model is specified with a MODEL statement just as
- in other SAS regression analysis procedures. The reason for
- using PROC TSCSREG instead of other SAS regression procedures is
- to be able to incorporate a model for the structure of the random
- errors.
-
- ( Here we put the files on our mainframe for downloading.)
-
- SASTSCSR.EXE is the executable file containing the TSCSREG
- procedure. Once downloaded, the TSCSREG procedure can be
- installed by copying SASTSCSR.EXE into your SASSASEXEETS
- subdirectory.
-
- In addition, these new DATA step functions are available:
-
- * UWUAIRY.EXE the airy function
- * UWUCNONC.EXE Chi-square noncentrality function
- * UWUDAIRY.EXE derivative of the airy function
- * UWUIBESS.EXE modified bessel function
- * UWUJBESS.EXE Bessel function
- * UWUPROBB.EXE bivariate normal function
- * UWUTNONC.EXE t noncentrality function
- * UWUFNONC.EXE f noncentrality function ,
-
- To install these new DATA step functions, copy these executable
- modules to the CORE subdirectory that is referenced in your
- CONFIG.SAS file.
-
- TSCSREG1.SAS, TSCSREG1.LST and TSCSREG1.LOG are the sample
- program, resulting listing and log files for the TSCSREG
- procedure. These files can be viewed and/or downloaded also.
- Preliminary documentation is available from SI anyone wanting to
- use this proc will need this documentation.
-