Merge pull request #296 from andersonfrailey/removefiler

Replace filer variable with data_source
PSLmodels · Jan 10, 2019 · 9bbe5d0 · 9bbe5d0
2 parents cb0c6cd + 4b06823
commit 9bbe5d0
Show file tree

Hide file tree

Showing 6 changed files with 7 additions and 6 deletions.
diff --git a/cps_data/cps.csv.gz b/cps_data/cps.csv.gz
diff --git a/cps_data/finalprep.py b/cps_data/finalprep.py
@@ -181,7 +181,7 @@ def drop_vars(data):
         'e07400', 'e07600', 'e09700', 'e09800', 'e09900', 'e11200', 'e17500',
         'e18400', 'e18500', 'e19200', 'e19800', 'e20100', 'e20400', 'g20500',
         'e24515', 'e24518', 'e26270', 'e27200', 'e32800', 'e58990', 'e62900',
-        'e87530', 'elderly_dependents', 'f2441', 'f6251', 'filer', 'n24',
+        'e87530', 'elderly_dependents', 'f2441', 'f6251', 'n24',
         'nu05', 'nu13', 'nu18', 'n1820', 'n21', 'p08000', 'p22250', 'p23250',
         'p25470', 'p87521', 's006', 'e03210', 'ssi_ben', 'snap_ben',
         'vet_ben', 'mcare_ben', 'mcaid_ben', 'oasdi_ben', 'other_ben',

diff --git a/puf_data/finalprep.py b/puf_data/finalprep.py
@@ -76,6 +76,8 @@ def main():
     # - Impute pension contributions:
     data = impute_pension_contributions(data.copy())
 
+    # Rename 'filer' to 'data_source'
+    data = data.rename(columns={'filer': 'data_source'})
     # - Write processed data to the final CSV-formatted file:
     if BENPUF:
         data.to_csv('benpuf.csv', index=False)

diff --git a/tests/cps_agg_expected.txt b/tests/cps_agg_expected.txt
@@ -47,7 +47,6 @@ e32800                     28913668              0           1666
 elderly_dependents             1724              0              3
 f2441                        163487              0             10
 ffpos                        498582              1             11
-filer                        437439              0              1
 fips                       12469516              1             56
 h_seq                   22850538907              1          99461
 housing_ben                60379435              0          53253

diff --git a/tests/puf_agg_expected.txt b/tests/puf_agg_expected.txt
@@ -10,6 +10,7 @@ age_head                   11691928              1             85
 age_spouse                  6247637              0             97
 agi_bin                     1890431              0             18
 cmbtp                    4316456617      -23842830       83355100
+data_source                  241245              0              1
 e00200                  39430402348              0       56530000
 e00200p                 23634568315              0       28265000
 e00200s                 15795834033              0       28265000
@@ -71,7 +72,6 @@ e87530                     22144920              0         147200
 elderly_dependents             1236              0              3
 f2441                         16109              0              3
 f6251                         55220              0              1
-filer                        241245              0              1
 fips                        6743572              1             56
 g20500                     16670714              0         720000
 k1bx14p                 -2197846801      -18990000        2341800

diff --git a/tests/records_metadata.json b/tests/records_metadata.json
@@ -922,13 +922,13 @@
       "max": 11
     }
   },
-  "filer": {
+  "data_source": {
     "type": "int",
-    "desc": "1 if unit files an income tax return; 0 if not (not used in tax-calculation logic)",
+    "desc": "1 if unit is created primarily from IRS-SOI PUF data; 0 if created primarily from CPS data (not used in tax-calculation logic)",
     "form": {
       "2013-2016": "sample construction info"
     },
-    "availability": "taxdata_puf, taxdata_cps",
+    "availability": "taxdata_puf",
     "range": {
       "min": 0,
       "max": 1