This example demonstrates how to request the CBI time series, which show the change in private industries for three different vintage dates: 1947-08-17, 1966-08-11, and 1994-08-26. Using the early range of START=’1942-01-01’and END=’1947-04-01’, you can get an idea of how the changes show up for each vintage date. If you specify OUTPUT=2, each time series is named by concatenating the series ID to the vintage date with an underscore.
title 'Specifying One Series ID with Multiple Vintage Dates for OUTPUT=2 Option'; libname _all_ clear; libname fred sasefred "%sysget(FRED)" OUTXML=fredex05 AUTOMAP=replace MAPREF=MyMap XMLMAP="%sysget(FRED)fredex05.map" APIKEY='XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX' IDLIST='CBI' VINTAGE='1947-08-17,1966-08-11,1994-08-26' START='1942-01-01' END='1947-04-01' FREQ='q' OUTPUT=2 UNITS='lin' FORMAT=xml; data threeVinsCBI; set fred.fredex05; run;
proc contents data=threeVinsCBI; run; proc print data=threeVinsCBI; run;
Output 39.5.1: Specifying One Series ID with Multiple Vintage Dates for OUTPUT=2 Option
Specifying One Series ID with Multiple Vintage Dates for OUTPUT=2 Option |
Obs | date | CBI_19470817 | CBI_19660811 | CBI_19940826 |
---|---|---|---|---|
1 | 1942-01-01 | 3.9 | . | . |
2 | 1942-04-01 | 3.6 | . | . |
3 | 1942-07-01 | -0.9 | . | . |
4 | 1942-10-01 | -0.9 | . | . |
5 | 1943-01-01 | -2.4 | . | . |
6 | 1943-04-01 | -2.1 | . | . |
7 | 1943-07-01 | 1.1 | . | . |
8 | 1943-10-01 | -1.5 | . | . |
9 | 1944-01-01 | -2.4 | . | . |
10 | 1944-04-01 | -3.2 | . | . |
11 | 1944-07-01 | -1.0 | . | . |
12 | 1944-10-01 | -1.3 | . | . |
13 | 1945-01-01 | -2.8 | . | . |
14 | 1945-04-01 | -1.5 | . | . |
15 | 1945-07-01 | 0.1 | . | . |
16 | 1945-10-01 | -0.8 | . | . |
17 | 1946-01-01 | 2.3 | 5.9 | 5.7 |
18 | 1946-04-01 | 2.0 | 8.8 | 8.6 |
19 | 1946-07-01 | 4.9 | 6.1 | 5.9 |
20 | 1946-10-01 | 5.4 | 4.7 | 4.5 |
21 | 1947-01-01 | 2.7 | 0.4 | 0.4 |
22 | 1947-04-01 | 1.5 | -1.0 | -1.2 |