Help using Base SAS procedures

FYI: USEFUL SAS MACRO CODE FOR INDUSTRY CLASSIFICATION INTO 15 INDUSTRIES

Reply
N/A
Posts: 1

FYI: USEFUL SAS MACRO CODE FOR INDUSTRY CLASSIFICATION INTO 15 INDUSTRIES

************************************************************************************************************************;

* SAS MACRO CODE FOR INDUSTRY CLASSIFICATION INTO 15 INDUSTRIES (prepared by Yaniv Konchitchki, U.C. Berkeley).

The code is based on Konchitchki, Yaniv. 2011."Inflation and Nominal Financial Reporting:

Implications for Performance and Stock Prices.” The Accounting Review 86 (3), 1045–1085.

This classification is also used in Barth, Mary E., Yaniv Konchitchki, Wayne R. Landsman. 2013. “Cost of

Capital and Earnings Transparency.” Journal of Accounting and Economics 55 (2-3), 206–224.

Please cite if using it.  This code creates a dataset "dataout" by adding to any existing dataset "datain" the

15-industry classification based on "varForSICcode".  The new dataset adds to your dataset an industry variable

denoted as specified in "varForIndDescOutput".

A suggested use:

%Add15KonchitchkiIndustries(datain=cstat, dataout=cstat, varForSICcode=sic, varForIndDescOutput=ind)

Note: The variable SIC is directly from Compustat (used to be DNUM in the historical Compustat legacy format);

%macro Add15KonchitchkiIndustries(datain=,dataout=,varForSICcode=,varForIndDescOutput=);

data &dataout;

  set &datain;

  varForSICcodeNum = &varForSICcode*1;* Ensure that the input industry classification is a numeric variable;

       if (1000<=varForSICcodeNum< 1300) or (1399< varForSICcodeNum<=1999) then &varForIndDescOutput = "Mining, constructi.";

  else if (2000<=varForSICcodeNum<=2111)     then &varForIndDescOutput = "Food";

  else if (2200<=varForSICcodeNum<=2799)     then &varForIndDescOutput = "Textiles, printing.";

  else if (2800<=varForSICcodeNum<=2824) or (2840<=varForSICcodeNum<=2899) then &varForIndDescOutput = "Chemicals";

  else if (2830<=varForSICcodeNum<=2836)     then &varForIndDescOutput = "Pharmaceuticals";

  else if (2900<=varForSICcodeNum<=2999) or (1300<=varForSICcodeNum<=1399) then &varForIndDescOutput = "Extractive Industr.";

  else if (3000<=varForSICcodeNum< 3570) or (3579< varForSICcodeNum< 3670)

  or (3679< varForSICcodeNum<=3999) then &varForIndDescOutput = "Durable";

  else if (7370<=varForSICcodeNum<=7379) or (3570<=varForSICcodeNum<=3579)

                                         or (3670<=varForSICcodeNum<=3679) then &varForIndDescOutput = "Computers";

  else if (4000<=varForSICcodeNum<=4899)     then &varForIndDescOutput = "Transportation";

  else if (4900<=varForSICcodeNum<=4999)   then &varForIndDescOutput = "Utilities";

  else if (5000<=varForSICcodeNum<=5999)   then &varForIndDescOutput = "Retail";

  else if (6000<=varForSICcodeNum<=6411)   then &varForIndDescOutput = "Financial Institut.";

  else if (6500<=varForSICcodeNum<=6999)   then &varForIndDescOutput = "Insurance, real es.";

  else if (7000<=varForSICcodeNum< 7370) or (7379< varForSICcodeNum<=8999) then &varForIndDescOutput = "Services";

  else if (9000<=varForSICcodeNum) then &varForIndDescOutput = "Other";

  else &varForIndDescOutput = "Other";

  drop varForSICcodeNum;

  run;

%mend Add15KonchitchkiIndustries;

************************************************************************************************************************;

Attachment
Attachment
Super User
Posts: 11,343

Re: FYI: USEFUL SAS MACRO CODE FOR INDUSTRY CLASSIFICATION INTO 15 INDUSTRIES

Posted in reply to CodingMaster

Or use a format such as:

proc format library=work;

value SICCode

0001 - 1300, 1399 - 1999 = "Mining, constructi."

2000 - 2111 = "Food"

2200 - 2799 = "Textiles, printing."

2800 - 2824, 2840 - 2899  = "Chemicals"

2830 - 2836  = "Pharmaceuticals"

2900 - 2999, 1300 - 1399 = "Extractive Industr."

3000 - 3570, 3579 <-< 3670, 3679 <- 3999 = "Durable"

7370 - 7379, 3570 - 3579, 3670 - 3679 = "Computers"

4000 - 4899 = "Transportation"

4900 - 4999 = "Utilities"

5000 - 5999 = "Retail"

6000 - 6411 = "Financial Institut."

6500 - 6999 = "Insurance, real es."

7000 -< 7370, 7379 <- 8999 = "Services"

9000 <- HIGH = "Other"

other = "Other";

run;

Assign the format to the SIC variable in analysis or print.

Ask a Question
Discussion stats
  • 1 reply
  • 725 views
  • 0 likes
  • 2 in conversation