User manual SAS SERVICE PARTS OPTIMIZATION ADAPTER 4.2 FOR SAP

[. . . ] For a hard-copy book: No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, or otherwise, without the prior written permission of the publisher, SAS Institute Inc. For a Web download or e-book: Your use of this publication shall be governed by the terms established by the vendor at the time you acquire this publication. Government Restricted Rights Notice: Use, duplication, or disclosure of this software and related documentation by the U. S. government is subject to the Agreement with SAS Institute and the restrictions set forth in FAR 52. 227-19, Commercial Computer Software-Restricted Rights (June 1987). [. . . ] A superscript of 2 (Yes2 or No2 ) indicates an option that works differently for in-database computing compared to out-of-database computing. A superscript of 3 (Yes3 , No3 , N/A3 , or Sometimes3 ) indicates an option that causes the procedure to use SAS/ACCESS for out-of-database computing. A superscript of 4 (Yes4 or No4 ) indicates an option that neither works for in-database computing nor prevents in-database computing, but is unlikely to cause a serious error. See the Comment column for more information. A superscript of 5 (Yes5 or No5 ) indicates an option that might cause incorrect answers or Teradata syntax errors. NOTE: As a data set option, DATABASE= is an alias for SCHEMA=, but as LIBNAME options, DATABASE= and SCHEMA= are distinct options. The data set options DATABASE= and SCHEMA= work the same way as the LIBNAME option SCHEMA=. If you specify the same option as both a data set option and as a LIBNAME option, the value of the data set option is used for that data set. Table 1. 5 SAS/ACCESS Data Set Options for Teradata Data Set Option BL_CONTROL= BL_DATAFILE= DATA= N/A N/A OUT= N/A N/A Comments For out-of-database I/O. Chapter 1: SAS Analytics Accelerator 1. 3 for Teradata: Guide Data Set Option BL_LOG= BUFFERS= BULKLOAD= CAST= CAST_OVERHEAD_MAXPERCENT= DATABASE= DATA= N/A N/A N/A N/A N/A No5 OUT= N/A N/A N/A N/A N/A No5 Comments For out-of-database I/O. DATABASE= is an alias for SCHEMA= as a data set option. DATABASE= as a data set option does not work like DATABASE= as a LIBNAME option. For SAS data sets only. BUFNO= BUFSIZE= CNTLLEV= COMPRESS= DLDMGACTION= DROP= ENCODING= N/A N/A N/A N/A N/A Yes1 No2 N/A N/A N/A N/A N/A No3 No2 36 !Chapter 1: SAS Analytics Accelerator 1. 3 for Teradata: Guide Data set option ENCRYPT= FILECLOSE= FIRSTOBS= DATA= No2 N/A No3 OUT= No2 N/A No3 Comments Requires SAS passwords. DBMS tables have no inherent row order, and FIRSTOBS= is defined in terms of row order. Neither the Teradata nor TSSQL drivers support generation options that are explicitly specified in the procedure statement, and the procedure does not know whether a generation number is explicit or implicit. Neither the Teradata nor TSSQL drivers support generation options that are explicitly specified in the procedure statement, and the procedure does not know whether a generation number is explicit or implicit. For SAS data sets only. GENMAX= No2 No2 GENNUM= No2 No2 IDXNAME= IDXWHERE= IN= INDEX= KEEP= LABEL= N/A N/A N/A N/A2 Yes1 No2 N/A N/A N/A No2 No3 No2 SAS data set attributes cannot be stored in Teradata. However, LABEL= should work on DATA=, just as TYPE= works on DATA=. DBMS tables have no inherent row order, and FIRSTOBS= is defined in terms of row order. SAS passwords are not supported on DBMSs. OBS= No3 No3 OBSBUF= OUTREP= POINTOBS= PW= N/A N/A N/A No2 N/A N/A N/A No2 Deploying and Using SAS Formats in Teradata !37 Data set option PWREQ= DATA= No2 OUT= No2 Comments SAS passwords are not supported on DBMSs. SAS passwords are not supported on DBMSs. [. . . ] You must be careful when formatting your SQL code for in-database processing. Proc SQL noerrorstop; connect to &dbms. (&connopt. ); execute ( create view tera. tdview_reg as select age, amount, checking, coapp, depends, duration, employed, existcr, 'foreign', history, housing, installp, job, marital, other, property, purpose, resident, savings, telephon, case when good_bad = 'good' then 0 else 1 end as target from tera. dmagecr) by &dbms; execute (commit) by &dbms. ; run; 7 Finally, you build a regression model. Even though the conversion of a binary character dependent vari- able to a numeric variable for use in a linear regression model is not optimal, it is a common operation for using the efficient REG procedure for model selection and exploratory modeling steps. [. . . ]