User and system description of SAS2ARGUS A SAS macro to execute τ
Contents
1. The following methods can be used for secondary suppression Description GH TabNo AprioriBoundPercentage ModelSize GH miter or as it is usually called The Hypercube Method Together with NET the only methods available in t ARGUS without access to commercial optimizers The secondary suppress mechanically without regard to the optimization and is not recommended as a method since it often causes great loss of information in the table MOD TabNo MaxTimePerSubtable A partial method that breaks down a hierarchical table of several non hierarchical tables protects them and finally composes a fully protected hierarchical table With the MaxTimePerSubtable parameter it is possible to limit the optimizer to work with each subtabell Is indicated in minutes OPT TabNo MaxComputingTime A method that protects a hierarchical table without breaking it down into smaller tables By setting the maximum time it is possible to limit the time for the optimizer to work Anges i minuter Indicated in minutes NET TabNo Network Solutions to be used for large two dimensional tables with a hierarchy Requires some special circumstances but no optimizer 2012 05 11 09 20 Document Page Statistiska centralbyran Statistics Sweden SCB IT AS Anders Kraftling SAS2Argus user manual 13 28 Date Version 2011 09 06 1 0 SUPPRESS RND Code ajwa Description cont RND TabNo RoundingBase Steps 1 Time Partitions St2. in the t ARGUS log Tip Set parameter debug 1 then the log from t ARGUS is included in the SAS log The parameter SafetyRule can take many arguments and defines the primary suppression or identification of primary unsafe cells Note that several primary suppression principles can be chosen these are then separated by the The following rules can be set P NK ZERO FREQ REQ WGT MIS and described in the table below 2012 05 11 09 20 Document Page ALIA Statistiska centralbyr n Statistics Sweden SCB IT AS Anders Kraftling SAS2Argus user manual 11 28 Date Version 2011 09 06 1 0 SAFETYRULE P NK ZERO FREQ REQ MIS WGT MAN All rules can occur multiple times separated by Additional arguments are typically set in subsequent brackets Description Percent Rule where additional arguments specified as P p n where nis optional and the default set to 1 P 20 3 represent percentage rule where p 20 and n 3 Dominance Rule where additional argument is specified as NK n k where n represents the number of items that may not account for more than k percent of the contribution of the cell domain Margin of zero cells in which additional argument sets the ZERO ZeroSafetyRange and refer to the size of the margin The Frequency Rule in with additional argument is specified as FREQ MinFreg FrequencySafet yRange where MinFreq specifies the minimum acceptable frequency and Frequency
3. or less observations contribute for 70 percent of the cells value the cell is regarded as uncertain 2012 05 11 09 20 Document Page Statistiska centralbyran Statistics Sweden SCB IT AS Anders Kraftling SAS2Argus user manual 12 28 Date Version 2011 09 06 1 0 SUPPRESS GH MOD OPT NET e or FREQ 3 30 implies that if the frequency is less than 3 and an intruder who knows his own contribution in the cell should not be not able to reveal another contributors value by less than 30 percent margin In that case the cell is considered to be uncertain e orMIS 1 means that if any contribution to the cell is missing the cell can be considered safe The first two P and NK is then assumed to be applied to the individual level and the subsequent P and NK is then assumed to be applied to the group level managed defined by the parameter Holding The first FREQ and REQ is assumed to be applied for the individual level and subsequent to the consolidated level ZERO can be entered only once for each security rule 2 4 Secondary Suppression SUPPRESS Here the arguments that can be used for parameter Suppress is described It has the same syntax as the SafetyRule that is additional arguments are given in a subsequent brackets Note that the first of these additional parameters specify the table number TabNo and as the SAS macro only handles one table at a time then this additional parameter always be 1
4. 2 2 Hierarchies How to describe hierarchies may need to be explained If one has to deal with hierarchical data these can be of two types e Levels in a value domain e Hierarchies consisting of merger of different values The first is exemplified by the Counties municipalities parishes in a six digit code that can be said to describe level 1 in the first two positions level 2 in the next two positions and finally level three in the last two positions The second is exemplified by specifying for example the county which forms a country region in list form How such a list could look like is described in the manual of t ARGUS and the text file Hierarchy file HRC must be created manually However when established the filename can be specified as an addition to the parameter Explanatory In the macro SAS2Argus one can describe the first type of hierarchy as follows Alt 1 Region 2 2 2 If the case of county municipality parish The concept of t ARGUS for this is lt HIERLEVELS gt The other option available is to specify a file name as follows Alt 2 Region Region hre If the hierarchy is described in a file named Region hrc The second argument within the parentheses which is an option in the option represents the character to be used as so called lead string The concept within t ARGUS for this is lt HIERCODELIST gt and lt HIERLEADSTRING gt respectively See the t ARGUS manual to better
5. 2012 05 11 09 20 Document Page SAS2Argus user manual 10 28 ALIA Statistiska centralbyr n l Statistics Sweden Date Version SCB IT AS Anders Kraftling 2011 09 06 1 0 understand how hierarchies are handled if this would be the case in your context There is much we don t have to write to get a metadata description of the hierarchies as we can deduce the following from these two types of options given together with the arguments lt HIERARCHICAL gt The name of the variable that is hierarchical and either option 1 lt HIERLEVELS gt The grouping of a string Or option 2 lt HIERCODELIST gt The name of the file that defines the hierarchy lt HIERLEADSTRING gt Special characters to interpret in the file 2 3 Risk assessment SAFETY RULE In a disclosure context it could be considered as relatively easy to define which cells domains that are unsafe to publish If it is possible to formulate a rule for which cells domains that are not safe to publish these cells domains are fairly easy to identify with relatively simple tools technology and to also suppress In t ARGUS the risk assessment rule is specified in the parameter SafetyRule The same concept word is used within the SAS macro The argument given here should be given in the style described in t ARGUS manual There is no check of the specified arguments done by the SAS macro Information about any inaccuracies in the arguments can then be seen
6. 97 61 66 79 97 174 2230 116 61 34 47 74 104 2699 542 163 93 83 119 126 2793 2288 278 64 44 35 23 8605 805 364 285 317 435 569 Ensamst ende f r ldrar 1212 191 58 43 57 70 96 28 2 10 5 3 7 1 139 16 5 8 14 15 22 415 50 19 16 18 26 28 404 33 8 8 14 17 36 150 27 10 2 7 4 9 76 63 6 4 1 1 0 1136 128 52 39 56 69 96 In this scenario you need to restructure the data file so that the cells can be identified by its properties We then get the entire table in list form where each cell is represented on one row 2012 05 11 09 20 Document Page SAS2Argus user manual 20 28 siB Statistiska centralbyr n Statistics Sweden Date Version SCB IT AS Anders Kraftling 2011 09 06 1 0 radtext Sex Fam Age Ink Resp Summa kvinnorochman T T T T 21241 Summa kvinnor och man T T T kol5 6804 Summa kvinnor ochm n T T T kol6 1307 a Summa kvinnor och man T T T kol 738 5 Summa kvinnor och man T T T kol8 786 6 Summa kvinnor och man T T T kol9 916 Summa kvinnor och m n T T T kol10 1152 8 Summakvinnorochman T T T kal11 1673 9 Summakvinnorochman T T T kol12 2416 Summa kvinnor och m n T T T koll3 1770 11 Summa kvinnor och man T T T koll4 1291 Summa kvinnor och man T T T kol15 883 13 Summa kvinnor och man T T T koll6 963 14 Summa kvinnor och man T T T koll 273 Summa kvinnor och man T T T koll8 269 16 daray 20 24 r T T 20 24 T 1398 d rav 20 24 r T T 20_24 kol5 199 Radtext is in the above example although redunda
7. Check conformity between designated variable roles and actual variables in the dataset eee oO en a me Variable_Meta amp _dataset oe ee eee Create the textfiles RDA meta and CSV data according to the information established so far and found in_Variable_Meta peed eed cea ae ea de at ek E Write_Datafile Dataset bquote amp _dataset Datafile amp PATH_tmp amp jobname oo ee Create the Command file BAT for t Argus as a manifest of what we want t Argus to do suffix ARB er Ss eee eee eee et ees see ee Oe See See Write_Jobfile Jobfile amp PATH_tmp amp jobname ARB end ice If debug 1 then include the ARB file in the SAS log beeen ae a Se i op aaa So eee sei if amp debug then do Fetch_Arb Arbfile amp PATH_tmp amp jobname ARB end ae Te Execute t Argus shee eet Nd ee oe eee if amp RunArgus then do fears ay ee OEE Remove the LOG for this job E RUCEEE SUE Re VE SEO ROSE UT SUMECUEND EN Me auau ae Remove_File YF CE PLCC ANE Ea 7 AP IE Ras Ses NEON Scere Execute t Argus eh ene em nr ne ae eee data _null_ _cmd amp PATH_exe amp PATH_tmp amp jobname ARB call system _cmd run 2012 05 11 09 20 CUI Statistiska centralbyr n l u Statistics Sweden SCB IT AS Anders Kraftling Document Page SAS2Argus user manual 28 28 Date Version 2011 09 06 1 0 If SAS 1 then import both TEXT files and HTML files produced by t Argus to this SAS s
8. for both microdata and aggregated data Explanatory Specifies the name names of the so called explanatory variables or dimensional Response Shadow Cost Lambda variables that spans the table Must be defined Along with the argument two options has been implemented If the variable is hierarchical one can ina subsequent brackets add a description of how it is hierarchical or in which text file that description can be found Note that variable names specified with spaces as separators if there is more than one in a list See Section 2 2 Hierarchies where this is described This specifies the name of response variable Must be defined The name of any shadow variable A company s turnover could be such a help variable If not specified then t ARGUS uses the Response variable Identification of potential cost variable This is a weight that can be used when respondents has given permission in advance to publish their values by putting a high cost on observations with permission If not specified then t ARGUS uses the Response variable Transformation Parameters used in a simplified Box Cox function as an exponent of the cost COST Default 1 Variables and their roles specific for aggregated data Frequency LowerLevel UpperLevel MaxScore Status TotCode The name of the variable that describes the frequency Must be defined for aggregated data Otherwise peculiar result can be produced since t ARGUS tries to c
9. of the rules specified in the parameter safety rule but can of course created manually as well Variables and their roles the specific micro data Weight Holding Request The name of the variable that contains weight The name of the variable that contains information about the corporate group When observations belonging to the same corporate group should be grouped together in the input file The name of the variable that indicates the status when the respondent has requested protection of data or not Inverse of consent If consent is relevant see commentary on status Holding and Request is the result of user requirements and the most widely used in Business Statistics and Foreign Trade Statistics Selection of output from t Argus Out Possible choices TABLE gt VarName delimiter Primary x Secondary PIVOT 0 gt VarName No Status PIVOT 1 gt VarName Status CODE 0 gt NoName delimiter Primary Secondary x No Status CODE 1 gt NoName delimiter Primary Part Secondary x No Status CODE 2 gt NoName delimiter Primary status 1 5 11 14 CODE 3 gt NoName delimiter Primary Part Status 1 11 SBS gt NoName delimiter Exp 0 Exp 0 zero deleted Status V D A zero deleted Status V D A INTER 0 gt NoName delimiter status only S M U INTER 1 gt NoName delimiter Status S M U If the parameter SAS 1 then we import PIV
10. path to the SAS log SS SS a a a re a eS es a a ee a ee eS sSystem_parameters Here ends the initialisation of the SAS session for use of the bridge between SAS and tau Argus A 2012 05 11 09 20 Document Page SAS2Argus user manual 16 28 SEB Statistiska centralbyran Statistics Sweden Date Version SCB IT AS Anders Kraftling 2011 09 06 1 0 After the initialization of the general part follows the specific part that preferably begins by defining a 1ibname to any data source database so that we can access data Here beneth follows the specific part the production part of the concept a Libname to data Bf E pp a a Pa mp ee a libname APPdat amp PATH app APPdat Note that 1ibname can also be an OLEDB connection to SQL server 3 1 SAS2Argus Examples of syntax Here are some examples of syntax which calls the macro 3 1 1 Aggregated Table Here is an example of an aggregated table InTable which risk is assessed by the frequency rule FREQ 5 30 meaning that there must be at least 4 observations in each cell accounting to 30 margin of safety EXAMPL 1 FREQUENCY Comment An aggregated table only primary suppressed Se ep ne a N gr sas2argus InTable APPdat TauFreq Jobname Examplel Explanatory sex fam age ink Frequency Resp SafetyRule FREQ 5 30 Out table pivot 1 code 3 inter
11. the SAS log Clean_Parameters sas Cleaning up parameters Check_Parameters sas Checks the parameter values Variable_Roles sas Sets variables roles Variable_Properties sas Sets variables properties Variable_Meta sas Checks that the specified variables exist in the specified data table SAS SQL Excel Write_Datafile sas Creates text file from table data Write_Jobfile sas Creates batch file Fetch_Arb sas Includes batch file in the SAS log Remove_File sas Removes any log file from previous runs Read_Datafile sas Imports text file from the t ARGUS to SAS Present_HTML sas Presents results report in SAS internal browser Argus2SAS sas Imports the output of t ARGUS to SAS by executing Read_Datafile sas and Present_HTML sas The inverse of the concept Fetch_Log sas Incorporates the log from t Argus in the SAS log Macro that controls the initialization of the SAS session System_Parameters sas A macro that controls the initial parameters required to set up the functionality in SAS Other Useful using macros utility macros Calculate_TopN sas A utility macro which aggregates PROC MEANS and presents the largest contributors in each cell Check_Dataset sas A utility macro to facilitate verification of the existence of a data set view table on SQL server Check_Outstring sas A utility macro that controls that the right arguments is specified for the output of t Argus Open_Editor sas A utility macro that opens
12. to be installed of course if Initial path location on disk EE pA EE E E EE ESEE E T E EEE SN EE slet PATH ini C SAS2Argus fk Path to the generic part and version and the application part Also give the full path to tau_Argus executable PATH exe Note Lt PATH Emp is nol Se gt we route to SAS Work as the working area Pnererable PATH tmp isi set auringi testi and enoublle shooummg jo be able to examine files ee ee if slet PATH sys amp Path_ini 1 1 slet PATH app amp Path_ini Demo slet PATH tmp amp Path_ini Demo APPtmp slet PATH exe C Program TauArgus tauARGUS exe There is a small program that initiates the path to the SAS programs The only general SAS programs used are A01 Init Session sas which in turn specifies the path to SAS macros according to the principle that is used in SAS by the AUTOSOURCE option This means that all macros are compiled in the time they are requested if Init the SAS session options notes nomprint nomlogic filename SASpgm amp PATH sys SASpgm S include SASpgm A01 Init Session sas Then we make one final check that everything is in place with a macro call Macro System Parameters Checks if full PATH to tau Argus exe is present so we can find it EXE file for tau Argus is present so we can use it PATH tmp is set explicit Otherwise we route to SAS WORK path and reports finally the used tau ARGUS work
13. via the macro 40 meaning that the max precision should be at least 40 for any intruder to be able to reveal a value of a suppressed cell against the remaining margin totals Hypercube method together with the NET method are the only methods available as a secondary suppression methods if you do not have access to an optimizer A purchased optimizer provides more options The Hypercube method puts out a lot of table cells in the secondary suppression phase as it works mechanically and is not recommended as first choice of method The Modular approach is better and to recommend as it tries to find the optimal solution that supress least secondary cells in the secondary suppression 3 1 4 Micro data Table Magnitude table A final example shows a micro data table that contains population and the county municipality parish region 2 2 2 represent one hierarchy and age classes Age 6 PATH _app APPdat Age hrc is the second hierarchy of classes In the latter case is the prepared file that describes the age group hierarchy in a file with a name along with the path specified as an argument in parentheses If HierLeadString not listed as here it is assumed to be Finally there is a third explanatory variable gender Sex The Response variable Count is in this case a column consisting of 1 EXAMPLE 3 HIERARCHY options nomprint nosource sas2argus InData APPdat Population Jobname E
14. 1 RunArgus 1 SAS 2 Debug 1 It is important for an aggregate table to indicate frequency as seen in the syntax above as it is essential information for t ARGUS when working with a pre aggregated table to do a risk assessment The naming of files based on Jobname used as a prefix to all files as part of an execution The following files are produced in the above call for t ARGUS Examplel csv The data file is tab delimited Examplel rda Metadata file Examplel arb Command file 2012 05 11 09 20 Document Page siB Statistiska centralbyr n Statistics Sweden SCB IT AS Anders Kraftling SAS2Argus user manual 17 28 Date Version 2011 09 06 1 0 There are also some helpful macros that facilitates the opening and to look at both input and output files Open_Editor opens any text file in the SAS Enhanced Editor Ssopen_editor Examplel out_5_1 rda Here in this example we open the file Examplel out _5 1 rda inthe SAS Enhanced Editor It is the metadata description for the named associated data file Examplel out _5 1 csv produced by t ARGUS From the name we can deduce that is a outputfil Type 5 1 from examplel That is the argument Inter 1 results in the 5th output format with the gross amount of information and status With the parameter SAS 2 then the result is returned imported to the SAS session and the produced data file can be found in the SAS Work library The R
15. 72 9 192 5 264 2 3093 642 349 361 470 592 931 1405 1098 9829 572 659 191 206 d rav 20 24 r 103 Gera 115 5 125 3 B 15 24 13 43 10 11 5 1 il 2 0 0 0 25 34 1239 96 6 211 7 219 1 GPa it Le Ri eee GHEE aeei IRA E 94 62 38 2 35 44 2334 95 8 272 5 2843 97 61 65 7S Sf 174 EE 4413 322 a A e 65 46 45 54 2230 94 8 294 1 310 2 116 61 34 47 740 104218 424 363 eea S178 228 69 90 55 64 2699 79 9 218 2728 542 163 93 83 4119 126 226 374 274 236 0 6 159 8 6192 48 64 65 2793 18 1 14 2 78 7 2288 278 64 44 35 23 20 12 8 6 3 6 2 4 20 64 8605 90 6 250 4 276 2 805 364 285 317 435 569 911 1393 1090 823 569 653 189 202 Ensamst ende f r ldrar 1212 84 2 206 5 245 2 191 58 43 5 70 96 145 212 134 74 45 59 13 15 d rav 20 24 ar 28 92 9 71 4 76 9 2 10 5 3 1 0 0 0 0 0 0 0 0 If table data for disclosure control initially looks like this where we for our example only have interest for the columns that are shaded in grey and holds information on the number of people in different income classes These values are oriented in the column headed from ko15 to ko118 radtext Tot kol5 kol kol kol8 kold koto Summa kvinnor och m n 21241 6804 1307 738 786 S16 1152 1398 199 206 178 169 175 144 2694 237 154 139 174 228 298 3648 318 124 11 119 158 245 3745 384 99 66 97 123 212 4292 1012 259 136 162 182 215 5464 4654 465 108 65 50 38 15777 2150 842 630 721 866 1114 11398 3093 642 349 361 470 592 103 8 15 24 13 13 10 1239 42 64 68 95 132 155 2334
16. Document Page Statistiska centralbyran Statistics Sweden SCB IT AS Anders Kraftling SAS2Argus user manual 1 28 Date Version 2011 09 06 1 0 User and system description of SAS2ARGUS A SAS macro to execute t Argus This is a description of a SAS macro SAS2Argus which is designed to facilitate risk assessment of tables and suppressing cells which are not able to publish regarding issues of integrity by establishing a bridge between SAS and t ARGUS User and system description Of SASZARGUS ssssssssssccssssssssccccecsssssssscesscesssssssssseeees 1 1 1 Recommended pedagogy csscccccccsssesssncccecesssesssneeeeceesseestsaeeeeeeesseeesiaeeeeseeeseenea 1 1 2 Choice of method rr 2 1 3 The purpose Of the SAS MACO wiiseccccccsssssssscceccesssessnneesececssessnsaeeeseeesseessaeesesessneeeaa 3 1 3 1 COMCE DTS iexssdesncissga ends ane i a siesta caauanedsatestecaaresesdssduainnsenuassdsscuaseass 3 1 4 The main macro SASZAIQUS csscccccccsssssssncceccesesessneeeeeeesseessuaeeeeeeesseeniaeeeeseseseesaa 4 2 SASZArgus Description cccccccccccsscccccssscccecsseccsccsseccsecessccsccssecccscssecccscsseccsscsssscsecssens 4 2 POVGIMGTOIS orena ck tice cdence ae sauidecsccen dee E puceaneandeceanaucteese ence sic E 4 2 2 Hier rehi Sirians araen aides tenavie di ua tds E titer RE EA a aE aa hastens 9 2 3 Risk assessment SAFETY RULE cccccccccsssssssceccecessssssneeeeecesssessneeessessseesnsaseeseeens 10 2 4 Sec
17. OT and INTER if choosen The easiest way may be to experiment in order to understand all the different options for output The most informative and useful may Intermediate considered to be Comment VarName means that variable names can be found in the file NoName means that no variable names are found in the text file Characters used as delimiters are listed in parentheses after Delimiter Characters that replaces primary suppressed values are given in parentheses after Primary Characters that replaces secondary suppressed values are given in parentheses after Secondary Status NoStatus indicates whether the cell status is reported in the output or not 2012 05 11 09 20 Document Page SAS2Argus user manual 9 28 ALIA Statistiska centralbyr n l Statistics Sweden Date Version SCB IT AS Anders Kraftling 2011 09 06 1 0 It is easy to understand that the amount of output format has a historical explanation as it is in the t ARGUS sphere are many stakeholders who had different preferences SBS is such a special format applied for Business Statistics Most formats is not able or are inappropriate to import into the SAS as a table They are best suited to present in tabular form in spread sheets as Excel The most informative and useful format is the Intermediate as it is possible to work interactively with i e edit in SAS and send back to t ARGUS for a second time and the intermediate is also easy to tabulate in SAS
18. Safet yRange indicates the safety margin for not be able to derive the suppressed frequency Request rule the request for confidentiality REQ Percentl1 Percent2 Safetymargin For example if an informant accounts for 70 and in that case have asked for protection This requires an additional variable in the table that indicates which items are requested not requested confidentiality with value 1 or 0 Variable name for the role is indicated by parameter Request Missing If MIS 0 which is the default means that the cells with a code for non response is still regarded as uncertain whether any SafetyRule is violated If MIS 1 then the cell i always considered as secure if at least one contributor has a missing value then it may be regarded that the contributor with miss cannot be identified In the SAS macro 9 has been defined as representative value for missing value If WGT O which is default then weights is not used when aggregating tables or in calculating the Safet yRule The name of the variable is indicated in the parameter Weight Manual safety margin default 20 This manually set safety margin is used only when the status is provided for each cell or when ana priori file is used to set the option that a particular cell is set manually to be uncertain Manual Unsafe i la Example K 3 70 FREQ 3 30 MIS 1 Explanation e NK 3 70 implies the dominance rule i e if 3
19. able The Frequency Resp is enough for to make an risk assessment based on just frequency System parameter option SAS 2 means that we also import importable data sets to SAS The so called intermediate table is obtained as output with the argument Out Inter 1 and as this table is suitable to import to SAS after the execution we import it and it will be found in SAS WORK and looks like this after execution Sex Fam Age Ink Resp Freqvar Resp_shadow Resp_cost Statusvar Lowerprotlevel Upperprotlevel 89 T IT 65_99 T 5464 5464 5464 5464 S 1 1 90 T T 65_99 kol5 4654 4654 4654 4654 S 1 1 91 T T 65_99 kol6 465 465 465 465 S 1 1 92 T IT 65_99 kol 108 108 108 108 1 1 93 T T 65_99 kol8 65 65 65 65 5 1 1 94 T IT 65 99 kol9 50 50 50 50 1 1 95 T JT 65_99 koll0 38 38 38 38 5 1 1 96 T JT 65_99 kol11 24 24 24 24S 1 1 97 T IT 65_99 koll2 20 20 20 20 1 1 98 T JT 65_99 koll3 12 12 12 125 1 1 99 T T 65_99 koll4 8 8 8 8s 1 1 100 T T 65_99 kol15 6 6 6 6S 1 1 101 T T 65_99 koll 6 6 6 6 6S 1 1 oom Tt T 65_99 koll 4 4 4 4u 1 6 1 6 103 T T 65_99 kol18 4 4 4 4u 1 6 1 6 Note the Statusvar with value S Safe and U Unsafe when frequency is below 5 The name of the output type Intermediate suggests by the name that this is the most appropriate format to work with as you can either edit and send it back to t Argus or forward it in the process chain to set up the presen
20. abular methods that modify values in an already derived table To attack an already made table by applying various suppression methods on table values are managed within the third category that is post tabular methods This is what the link between SAS and t ARGUS offers in its concept However there are situations when it is much easier to solve this problem and consider other solutions earlier in the process chain For example define the table with larger geographic areas to collapse groups and levels where the number of contributors are few The advantages are several The methods are easy to implement and provides understandable tables that are able to summaries and the technique is also easy to explain to the user Sometimes it is not possible if the definition of the tables are established in advance by i e international standard or that this approach gives too much information loss Pre tabular methods to adjust the values of micro data prior to table the request provide tables that you are able to summarize but it can be more difficult to describe for the user the used data adjustment methods 2012 05 11 09 20 Document Page SAS2Argus user manual 3 28 Statistiska centralbyran Statistics Sweden Date Version SCB IT AS Anders Kraftling 2011 09 06 1 0 Choice of method for protection of informers is still something that must be considered as it may significantly affect both the work involved the complexity
21. al information is written in the SAS log 1 Information is written to the SAS log and the log of t Argus is also included the SAS log default Facilitates debugging and documentation as all available information from execution is found in the SAS log Describes the macro and its parameters in the SAS log 0 No information in the log default 1 The macro is described in the log and the macro stops no execution The macro is indeed documented in the script code but this is an easy way to get access to a brief description An option that controls imports from t ARGUS to SAS 0 No import from t ARGUS to SAS default 1 Imports the results report in HTML format from t ARGUS and includes it in the SAS internal browser Also imports the output from the t Argus in what is called the intermediate format to the SAS WORK This is the only format suitable for import into a SAS dataset or table Parameters that define the input data into t ARGUS One of these may be selected either INPUT or INTABLE InData InTable Specifies the name of SAS data sets of micro data Note that this also can be an SQL table All data sources that SAS supports with access methods can be used Must be specified either InData or InTable Specifies the name of SAS data sets for already aggregated data Note that this also can be an SQL table All data sources that SAS supports with access methods can be used However the data must often be prepared in
22. any text file in the SAS program editor in SAS Open_Excel sas A utility macro that opens any CSV file in Excel from SAS 2012 05 11 09 20 Document Page r SAS2Argus user manual 26 28 ALIA Statistiska centralbyr n l Statistics Sweden Date Version SCB IT AS Anders Kraftling 2011 09 06 1 0 4 1 Context To describe the context it is easiest to in condensed form present the main macro SAS2Argus and how it in turn executes the underlying utility macros for those who want to understand the function from a system perspective macro SAS2Argus if amp help then do SAS2Argus_help end If parameter RUNARGUS is set to 2 gt Rerun t Argus with already created files meaning that there is no need to establish any CSV or RDA or ARB files This is the scenario when the user could changed the status on certain domains cells from SAFE to UNSAFE or from UNSAFE to SAFE and with a new request to t Argus to handle secondary cell suppression EES E in sta ee Re ee ae E if amp runargus ne 2 then do Ji EOE E E Establish the Metadata for the ROLES OF VARIABLES at macro invocation E E eg a E PO Establish the Metadata for the ACTUAL VARIABLES in the dataset given at macro invocation Variable_Properties amp _dataset 2012 05 11 09 20 Document Page ane SAS2Argus user manual 27 28 ALIA Statistiska centralbyr n yu Statistics Sweden Date Version SCB IT AS Anders Kraftling 2011 09 06 1 0 Phe
23. controls syntax generates all the necessary input files and start a batch job of t ARGUS The function of the macro is thus mainly to get SAS to establish the text files that ARGUS requires to be run through its appli cation interfaces API available in t Argus and the macro might also import the results to SAS after execution With an understanding of the macro and how the parameter is set and the function of t ARGUS it is relatively easy to perform risk assessment and or cell suppressing with use of the macro In place it is an effective workbench that facilitate the work significantly and gives the opportunity to test a variety of different parameters and methods The alternative is otherwise to execute t ARGUS through the graphical user interface GUI which may be a recommendation for a novice when it can provide a better understanding of all issues around disclosure in general and the features available in t Argus in particular But you will still need to fabricate the text files in the format and content in which t ARGUS expects to find them 2 SAS2Argus Description Here is a description of the macro SAS2Argus How to set up the SAS session to access the functionality of the SAS macro to reach the capa bilities of ARGUS We begin by first describing the parameters the macro utilize cause this in itself describes much of the potential of the concept 2 1 Parameters The macro is a so called Named style macro
24. e macro might look like with no other comments sas2argus InTable APPdat Freqdata Jobname Tab 1 Explanatory sex fam age ink Frequency Resp SafetyRule FREQ 5 30 Out inter 1 RunArgus 1 SAS 2 Debug 0 Note that the parameter description below you will also find some useful instructions Description General parameters system parameters JOBNAME RUNARGUS A name that is used as a prefix for all text files that are created for t ARGUS and used by t Argus in a job execution Default unless specified is SASZARGUS This makes it easier in that sense that you are able to see which files that belong together in an identifiable context An option that allows to control if 0 Only text files are created by the macro t Argus is not executed 1 Text files are created and t Argus is executed default 2 Don t create any text file execute only t ARGUS on already created text files This makes it possible to produce the text files first edit the text files manually and finally execute the manually edited text files To overcome exceptional situations not supported by the macro for example 2012 05 11 09 20 Document Page SAS2Argus user manual 6 28 Statistiska centralbyran Statistics Sweden Date Version SCB IT AS Anders Kraftling 2011 09 06 1 0 DEBUG HELP SAS An option for providing a way to get more information incorporated in the SAS log 0 No addition
25. ed and 2012 05 11 09 20 Document Page Statistiska centralbyran Statistics Sweden SCB IT AS Anders Kraftling SAS2Argus user manual 2 28 Date Version 2011 09 06 1 0 used in t Argus is also used in the SAS macro as far as possible to facilitate the understanding of how if fit together This description of the bridge between SAS and t Argus how ambitious it would be limp without the knowledge of the principles and methods used in t ARGUS A recommendation for a novice is to start in t Argus and get acquainted with the program and take advantage of the t ARGUS manual This smooth the progress significantly the understanding of the whole concept Are you also in need to get familiar and understand the implemented methods in order to choose the right we also recommended the methodological handbook manual where these methods are described 1 2 Choice of method There are a variety of methods to choose from in statistical disclosure control that can be used to produce privacy protected tables The choice of method depends on many factors such as how data is used how the method can be implemented from a practical perspective and what protection it finally gives the table when published Methods can be divided into three categories 1 Pre tabular methods to adjust the micro data before the table request 2 Table redesign methods that modify the design of the table by defining the level of detail 3 Post t
26. em eee nese n erase ewccnassasesseeces lt LOGBOOK gt C SAS24rgus Demo APPtmp Example1 LoG lt OPENTABLEDATA gt C SAS2Argus Demo APPtmp Example1 csy lt OPENHETADATA gt C SAS2A4rgus Demo APPtmp Example1 RDA lt SPECIFYTABLE gt Sex Fam Age Ink Resp Resp Resp 1 lt SAFETYRULE gt FREQ 5 40 lt READTABLE gt 1 lt WRITETABLE gt 1 1 0 C SAS24rgus Demo APPtmp Example1_Out_1_0 csv lt WRITETABLE gt 1 2 1 C SAS2Argus Demo APPtmp Example1_Out_2_1 csv lt WRITETABLE gt 1 3 3 C SAS24rgus Demo APPtmp Example1_Out_3_3 csv lt WRITETABLE gt 1 5 1 C SAS24rgus Demo APPtmp Example1_Out_5_1 csv NOTES Sass s essa aaa a aS a aaa aa aaa a a aa aS aS Sa SSS S SSeS 2 5 NOTE END of tau ARGUS ARB file NOTES Sa SsS SSS SSS aa Sa aaa Sa a See a ae ae a a ea aa a Sa aa aaa Se eee The log file from t Argus in also found written in the SAS log 3 1 2 Workflow The process of protecting the integrity of tables and the effort of work needed depends on the initial layout of the tables It can range from very simple work and already suitable oriented data to quite complicated that brings in work to reorganize the data in a way that is suitable for disclosure control A bit of where do we come from and where are we going Variables needed in the various roles in the elimination process must be available and each cell needs to be described by properties that has a unique ID that is the variab
27. ession Argus2SAS executes Read_Datafile A macro that reads delimited CSV text files with use of supplied metadata file RDA and establish SAS datasets Present_HTML A macro that presents the HTML file produced by t Argus in the SAS internal browser if amp SAS then do Argus2SAS end If debug 1 then include the HTML files from t Argus in the SAS internal browser and the LOG file from t Argus in the SAS log if amp debug then do Fetch_Log end end Y mend SAS2Argus All included macros are well commented in order to increase readability make the code function easier to understand and to facilitate system maintenance 2012 05 11 09 20
28. esult Report in the form of an HTML file from the t Argus can be found in SAS internal browser a bit depending on the settings of the SAS Ci Results Viewer SAS Output DER a Executed by SAS2ARGUS bridge between SAS and tau Argus SAS2ARGUS CASAS2Argus 1 1 SASmac T ARGUS Report Table created date 08 jun 2011 time 15 14 08 Original file CASAS2Argus Demo APPtmp Example1 CSV Meta file C SAS2Argus Demo APPtmp Example1 RDA Table file C SAS2Argus Demo APPtmp Example1_Out_3_3 csv Table read as table Table structure Explanatory var 1 Sex E xplanatory var 4 Ink Explanatory var 2 Explanatory var 3 Age sid Shadow variable Resp Cost variable Resp lt br gt Missing totals have been computed Safety Rule Minimum individual cell frequency 5 safety margin 40 2012 05 11 09 20 Statistics Sweden SEB i SCB IT AS Anders Kraftling Document Page Statistiska centralbyran SAS2Argus user manual 18 28 Date Version 2011 09 06 1 0 When parameter Debug 1 is set then the ARB file batch file is also written in the SAS log NOTE Macro FETCH_ARB NOTES 2 2 n nen n ene nnn e new e nen teen ene n mene nen eneeneeennannnseesenenans NOTE Here follows the established ARB file for tau ARGUS ee a ee i a ee a ee ee eee f 1 4 fp e a a This job based on dataset APPdat TauFreq f frre c crt c emcee ere e ete ese meee secant ne
29. les describing the dimensions need a unique value for each cell Other properties may be in excess of this weight cost shadow variable and so on If table data is not oriented in row dimension you have to restructure the data If we start from a table of income statistics which contains both a variety of statistical measures different dimension variables partly overlapping and a number of different income classes Initially a table could look like this 2012 05 11 09 20 Document Page SAS2Argus user manual 19 28 siB Statistiska centralbyr n Statistics Sweden Date Version SCB IT AS Anders Kraftling 2011 09 06 1 0 Andel Antal med radtext totalt inkomst Median Medel kol5 kol6 kol kol8 kol9 koll0 koll1 koll2 koll3 koll4 koll5 koll6 koll koll8 Summa kvinnor och m n 21241 68 167 6 246 7 6804 1307 738 786 916 1152 1673 2416 1770 1291 BOS Gd 273263 d rav 20 24 r 1398 85 8 120 6 140 6 199 206 178 169 175 144 131 98 44 35 17 2 0 0 25 34 2694 91 2 203 1 2227 237 154 139 174 228 298 348 400 294 178 132 89 11 12 35 44 3648 91 3 252 276 318 124 111 119 158 245 437 640 465 370 239 272 90 60 45 54 3745 89 7 267 2978 9384 99 Gb Sh 12s cee BS Tea 52d 352 T266 aaa 0d n 55 64 4292 76 4 202 2 2646 1012 259 136 162 182 215 350 571 427 348 223 8 260 64 83 165 5464 148 10 6 71 8 4654 465 108 65 50 38 24 20 12 8 6 6 4 4 20 64 15777 86 4 222 257 2150 842 630 721 866 1114 1649 2396 1758 1283 877 957 269 265 Gifta sambo 11398
30. ndidate for secondary suppression A high cost increases the likelihood of the same Two new values for the If the range for a particular cell in advance is likely to LowerLevel and be less than the normal range it may be stated here UpperLevel separated to get anew protection level by comma Note The possibility to change the status of a cell is of course limited This means that a cell that is primary unsafe should not be changed to protected Norcana protected cell is changed to unsafe 2012 05 11 09 20 L 1 LR Statistics Sweden SCB IT AS Anders Kraftling ALIA Statistiska centralbyr n Document Page SAS2Argus user manual 14 28 Date Version 2011 09 06 1 0 The cost COST must always be a positive value It is recommended to be restrictive to set cell status to protected Ifthe intention is to avoid the cell be subjected to a secondary suppression it is better to put a high cost for the cell If nevertheless would be secondary suppressed by the algorithm in t Argus there are good reasons for this See also the t ARGUS manual to get a description of file formats possible arguments and examples 2012 05 11 09 20 Document Page SAS2Argus user manual 15 28 siB Statistiska centralbyr n Statistics Sweden Date Version SCB IT AS Anders Kraftling 2011 09 06 1 0 3 SAS2Argus Usage To initiate the macro we only need a pointer to tell SAS where to find the resources t ARGUS have also
31. nt information for t ARGUS but otherwise the table is now usable as input to the disclosure process as set out above and used as input in our first example of a SAS2Argus call There are numerous techniques available to tweak the data and depends on the choice of tools SQL case pivot SAS transpose group data step We have now created dimensional variables from the only information available in radtext and have introduced a constant T that denotes the totals that is the table marginal totals The overlapping age classes 20 64 are not managed If we would handle this which would be an example of an hierarchy and then this fact also had to be described for t ARGUS If we now define a call to the macro to perform a risk assessment of the table with the criterion that there must be at least 5 observations in each cell in the line oriented data we now prepared We define the name of the dataset and the following roles rules and system parameters sas2argus InTable APPdat TauFreq Jobname Examplel Explanatory Sex Fam Age Ink Frequency Resp SafetyRule FREQ 5 40 Out inter 1 RunArgus 1 SAS 2 Debug 1 2012 05 11 09 20 SOB Statistiska centralbyran Statistics Sweden SCB IT AS Anders Kraftling Document Page SAS2Argus user manual 21 28 Date Version 2011 09 06 1 0 Note that for an aggregated frequency table we don t need to enter Response vari
32. ode not implemented within the macro There is another text file defined where you can set properties on the cells before secondary suppression is performed in a priori file The a priori file All of these are described in t ARGUS manual 1 3 1 Concepts To facilitate the mapping between the concepts used in the t ARGUS and concepts defined in the SAS macro SAS2Argus the same concepts have been used within the macro although some terms maybe are unusual for SAS users such as explanatory Similarly all the suffixes in the file types that are defined and referenced in t Argus is also used in the SAS macro to facilitate the reading of t ARGUS manual and understand the mapping between t Argus and SAS2Argus The macro consists of a main macro SAS2Argus and a set of utility macros that are called within the main macro The normal user should not really have to worry about these sub macros but are listed and described at The word used in t Argus for describing dimensions in the table 2012 05 11 09 20 Document Page SAS2Argus user manual 4 28 Statistiska centralbyran Statistics Sweden Date Version SCB IT AS Anders Kraftling 2011 09 06 1 0 the end of this document to some extent describe the components of main macros function 1 4 The main macro SAS2Argus The user needs only to understand the use of the comprehensive SAS macro SAS2Argus which then calls the execution of a sequence of other macros which
33. of the solution and the final effect 1 3 The purpose of the SAS macro The application interface in t Argus as used from SAS works with text files that must be created And that what s the SAS macro does Primary text files that need to be prepared to put t Argus in work are 1 Data file with either micro data or aggregated data CSV 2 A metadata file that describes this data file RDA 3 A batch file command file that describes the rules for the risk assessment and which type s of result file s to be produced The purpose of the macro is to facilitate the user to access the functionality of t ARGUS by automating the production of these necessary text files With the use of the metadata information that is available in a SAS data set or a SAS view against other data such as SQL tables much metadata information can be derived as data type number of decimal places and more Then the user needs only to add information about variables roles risk assessment method and prevention method In addition to these three primary text files additional text files need to be created to describe hierarchies in the data Hiearchy file HRC if applicable It is possible to define labels for values with text files Code List file CDL It is also possible to recode values in the data by use of text files When it is obvious simpler to do those tasks in SAS by defining labels and or recode data so is this possibility to label and rec
34. ompute the frequency The name of the variable indicating the lowest level of protection intervals The name of the variable indicating the highest level of protection intervals The name of variables that holds the single highest contributors in each cell Used in magnitude tables when the dominance rule is applied to pre aggregated tables The largest contributors can be computed with PROC MEANS There is a utility macro that can do this Calculate_TopN sas The name of any variable that indicates status Status value can then typically be S Safe U Unsafe P Protected Note that in the case of consent approval is not advisable to put the status indicator to S SAFE for cells observations with consent the price can be high with regard to secondary suppression of the table Use COST instead A constant that specifies indicates which value that represent the total of an aggregate table Default anv nds tecken T Default characteris T 2012 05 11 09 20 Document Page SAS2Argus user manual 8 28 Statistiska centralbyran Statistics Sweden Date Version SCB IT AS Anders Kraftling 2011 09 06 1 0 For aggregated data it is important not to forget to specify parameter frequency otherwise the t ARGUS trying to aggregate already aggrega ted data to determine the frequency of the number of contributions in each cell The Status variable is manufactured by t Argus when a risk assessment is made by use
35. ondary Suppression SUPPRESS cccccccccscsssessssecescceaseeessccesssssseesseceesseaauaes 12 24 1 lt A PrlOnt TG xs veedeet sce eseeicde tanadintesviason aces dues svtiesanesesd cee vsceeiavanxedvplestetiaesuenavnes 13 3 SASZArguS USAage ccssscsssssssssssssssssssssssssscscscssccscssssssssssssssssssssssssssssssssssssseees 15 3 1 SAS2Argus EXAMples Of SYNTAX oeeeccccccccsssssscecccecssssssneeeseceseseesseeeeeeesssessaaeeeeeess 16 3 11 Aperepated Table sicer rran ar EEA eso ees 16 3L WorkhOW ressens conrso naain ignei Nationa eia Eek ESEE EESAN AAE 18 3 1 3 Micro data Table Magnitude table ccccccccccccccsssseeeeceesesessseeeeeeeeseees 21 3 1 4 Micro data Table Magnitude table cccccccccccccccceeeeeeeeeeeeeeeeeeseeeeeeeeees 23 4 SAS2Arg s SUUCtUNG is sccccccossessecsvesscccuysserecsecensesacessrccneceadssaseeseccescssveceresseccavessesessee 25 AT 1COMLONE irie OERE E Aa E N EIE Cave eae cad Case wena 26 This description is a brief summary of the macro that makes it possible to put disclosure and cell value elimination in a process chain with use of SAS The subject is complex and data tables must often be prepared in a special way for it to be possible to protect in aspects of integrity of the table in a safe and rational way 1 1 Recommended pedagogy To understand the purpose and use of SAS macro SAS2Argus it is essential to know how the function of ARGUS how it is used Concepts defin
36. opRule RoundingBase refers to the basis for rounding Steps means the number of step allowed default 0 The fourth argument is a constant for future extension Partitions default 0 1 means the inverse StopRule default 3 1 Rapid only 2 First feasible solution 3 Optimal Solutions 2 4 1 A priori file An a priori file is a simple text file that can be created in an editor The information in A priori the file used after risk assessment but before secondary suppression The file updates including Status Cost LowerLevel and UpperLevel and can also put a new value on the level of individual cells when other knowledge of individual cells is at hand That is We are given an opportunity to bring existing knowledge to the table after the initial risk assessment process The current SAS macro has no parameter which in itself would be easy to add to deal with an a priori file but you can edit the batch file ARB by adding the argument lt APRIORI gt Filename TabNo Separator Each line begins with comma separated values that identify each individual cell and is followed after another comma with the following codes and values to provide information with a priori knowledge of specific cells Parameter Explanation Status changes to the safe Status changed to unsafe Status changed to protected New value for cost COST A low cost increases the likelihood that the cell becomes a ca
37. s into account within the specification when the metadata file is created Assign a SAS format for telling SAS2ARGUS how many decimals we should deal with when establishing the files for tau ARGUS SAE GT SS en ES Pe EON rt Ney me E S proc datasets library APPdat NoList modify TAUmicro agg format Resp 8 3 run quit Then we execute the main macro SAS2Argus EXAMPLE 2 Magnitude a I S a aSa options nonotes nosource nomprint nomlogic nosymbolgen sas2argus InTable APPdat TAUmicro_agg Jobname Example2 Explanatory Age Omk Respons Resp MaxScore Resp TopN_1 Resp TopN 2 Resp TopN 3 Resp TopN 4 SafetyRule NK 3 90 Suppress GH 1 40 0 Out table pivot 1 code 3 inter 1 RunArgus 1 SAS 2 Debug 1 2012 05 11 09 20 CUI Statistiska centralbyr n L LF Statistics Sweden SCB IT AS Anders Kraftling Document Page SAS2Argus user manual 23 28 Date Version 2011 09 06 1 0 Here we indicate beside Explanatory and Response also the with PROC MEANS produced MaxScore variables names The Safety rule is here NK 3 90 the so called NK rule stating that three companies may not contribute more than 90 of the contents of domain cell Secondary Suppression GH 1 40 0 is here done with the hypercube method or GH miter as it is called Where 1 stands for Table 1 and is set in general as we execute only one table at a time
38. some way and aggregate data must at least have information about the frequency in each cell to be useful as input to risk assessment and suppression Risk assessment and secondary suppression SafetyRule Specifies the method of risk assessment to be used This these arguments are not Suppress checked in a preventive way by the macro Study the t ARGUS manual for valid arguments Must be defined unless an the exception in the case cell status variable name Status is available and thus the risk assessment is already made See section 2 3 Risk assessment SAFETY RULE where this parameter is described Specifies the method for suppression to be used This these arguments are not checked in a preventive way by the macro Study the t ARGUS manual for valid arguments If this argument is omitted it means that only a risk assessment is done See Section 2 4 Secondary Suppression SUPPRESS where this parameter is described The information inthe safety rule and suppress ports in the command file ARB to tell t Argus what to do The rest of the information from the parameters below is used to create the data file 2012 05 11 09 20 Document Page SAS2Argus user manual 7 28 Statistiska centralbyran Statistics Sweden Date Version SCB IT AS Anders Kraftling 2011 09 06 1 0 CSV with the net content of required defined roles variables and metadata description of it RDA Variables and their roles Generic
39. tation tables 3 1 3 Micro data Table Magnitude table The previous example was an aggregated frequency table The example here is a magnitude table which we let SAS aggregate before we continue with the disclosure If we wish to apply the so called NK rule N objects may not contribute more than S of the cell contents we also need to identify the major contributors in each cell first 2012 05 11 09 20 Document Page SAS2Argus user manual 22 28 siB Statistiska centralbyr n Statistics Sweden Date Version SCB IT AS Anders Kraftling 2011 09 06 1 0 This can be accomplished using this syntax of PROC MEANS Summarize and select four TopN values for the explanatory variables proc means data APPdat TAUmicro Missing CompleteTypes NWay NoPrint class Age Omk var Resp output out APPdat TAUmicro_agg sum Resp idgroup max Resp out 4 Resp Resp_TopN NoInherit run where the four major donors will be written to the output dataset as told in the syntax Or if you have trouble remembering the syntax use the macro Calculate TopN which is one of utility macros Ji eae ee SS SE eee EEE E N E E E E E a ee E E See ee E ee ee Calculate_TopN Indata APPdat TAUmicro ClassVar Age Omk Var Resp TopN 4 By leaning against the metadata that we have available in SAS we can define a format to tell how many decimal places we want to work with Then the macro will see this and we get thi
40. that is it has a set of named parameters that are assigned values as arguments Note that the following description of the parameters is a gross set of the macro s possible parameters and that in practice only a few need to be specified in a typical scenario Most parameters describes roles for variables variables that also must exist or be established in advance for to later be able to reference those Some of these roles is unusual and rarely used A number of parameters are assigned default values if value is not specified explicitly Required parameters are marked in color 2012 05 11 09 20 Document Page SAS2Argus user manual 5 28 Statistiska centralbyran Statistics Sweden Date Version SCB IT AS Anders Kraftling 2011 09 06 1 0 Parameter The parameter list is divided into the following categories e General parameters system parameters e Parameters that define the input data into t ARGUS e Parameters for risk assessment and secondary suppression e Variables and their roles o General for both micro data and aggregated data o Specific for aggregated data o Specific for micro data e Selection choice of the output of t Argus As it is known as a Named style macro all parameters are separated by commas so that commas should not be used to separate for example listed variable names within a parameter To make the context of the parameters more understandable we here give a brief example of how an invocation of th
41. xample3 Explanatory Region 2 2 2 Age amp PATH_app APPdat Age hrc Sex Response Count SafetyRule NK 3 75 P 25 100 1 FREQ 5 30 Suppress GH 1 30 0 Out table inter 1 RunArgus 0 SAS 1 Debug 1 2 No commas in the list of numbers 2012 05 11 09 20 L LF Statistics Sweden SCB IT AS Anders Kraftling AULIA Statistiska centralbyr n Document Page SAS2Argus user manual 24 28 Date Version 2011 09 06 1 0 Several safety rules can be specified as seen See the t ARGUS manual for a detailed description of the arguments that s are possible RunArgus 0 means that only the text files that t Argus need to be produced t Argus is not executed These text files can be reused in the interactive interface SAS 1 has no effect here since t ARGUS will not be executed because of the previous arguments Would otherwise meant that the Result Report and the log had been imported and presented within the SAS session 2012 05 11 09 20 Document Page SAS2Argus user manual 25 28 Statistiska centralbyran Statistics Sweden Date Version SCB IT AS Anders Kraftling 2011 09 06 1 0 4 SAS2Argus Structure SAS2Argus is composed of a main macro that uses a set of utility macros Macro Description Sas2Argus sas The main macro which the user calls with parameter set and executes Macros that are called by the macro SAS2Argus SAS2Argus_Help sas Prints usage description of the macro to
Download Pdf Manuals
Related Search