Merge pull request #2740 from martinholmer/add-tmd

Add capability to handle tmd input

.github/workflows/build_and_test.yml

@@ -1,4 +1,4 @@
-name: Build Package and Test Source Code [Python 3.9, 3.10, 2.11]
+name: Build Package and Test Source Code [Python 3.9, 3.10, 3.11]
 
 on: [push, pull_request]
 
@@ -34,7 +34,7 @@ jobs:
         shell: bash -l {0}
         working-directory: ./
         run: |
-          pytest -m 'not requires_pufcsv and not pre_release and not local' --cov=./ --cov-report=xml
+          pytest -m 'not requires_pufcsv and not requires_tmdcsv and not pre_release and not local' --cov=./ --cov-report=xml
       - name: Upload coverage to Codecov
         if: matrix.os == 'ubuntu-latest'
         uses: codecov/codecov-action@v4

.gitignore

@@ -15,6 +15,7 @@ read-the-docs/_build/
 
 # IRS-SOI PUF data file
 puf.csv
+tmd.csv
 
 # Jupyter notebook checkpoints
 *.ipynb_checkpoints*

MANIFEST.in

@@ -7,4 +7,4 @@ include taxcalc/policy_current_law.json
 include taxcalc/puf_weights.csv.gz
 include taxcalc/puf_ratios.csv
 include taxcalc/records_variables.json
-
+include taxcalc/tmd_weights.csv.gz

Makefile

@@ -51,19 +51,19 @@ endef
 .PHONY=pytest-cps
 pytest-cps:
 	@$(pytest-setup)
-	@cd taxcalc ; pytest -n4 --disable-warnings -m "not requires_pufcsv and not pre_release"
+	@cd taxcalc ; pytest -n4 --disable-warnings --durations=0 --durations-min=2 -m "not requires_pufcsv and not requires_tmdcsv and not pre_release"
 	@$(pytest-cleanup)
 
 .PHONY=pytest
 pytest:
 	@$(pytest-setup)
-	@cd taxcalc ; pytest -n4 -m "not pre_release"
+	@cd taxcalc ; pytest -n4 --disable-warnings --durations=0 --durations-min=2 -m "not pre_release"
 	@$(pytest-cleanup)
 
 .PHONY=pytest-all
 pytest-all:
 	@$(pytest-setup)
-	@cd taxcalc ; pytest -n4 -m ""
+	@cd taxcalc ; pytest -n4 --disable-warnings --durations=0 --durations-min=2 -m ""
 	@$(pytest-cleanup)
 
 define tctest-cleanup
@@ -103,7 +103,7 @@ define coverage-cleanup
 rm -f .coverage htmlcov/*
 endef
 
-COVMARK = "not requires_pufcsv and not pre_release"
+COVMARK = "not requires_pufcsv and not requires_tmdcsv and not pre_release"
 
 OS := $(shell uname -s)
 

conda.recipe/meta.yaml

@@ -8,10 +8,10 @@ build:
 
 requirements:
   build:
-    - "python<3.12"
-    - "numpy>=1.14"
-    - "pandas>=1.2.0"
-    - "bokeh>=1.4.0, <3.0.0"
+    - "python>=3.9, <3.12"
+    - "numpy>=1.20, <2.0"
+    - "pandas>=2.0, <3.0"
+    - "bokeh>=1.4, <3.0"
     - requests
     - numba
     - "paramtools>=0.18.0"
@@ -21,10 +21,10 @@ requirements:
     - openpyxl
 
   run:
-    - "python<3.12"
-    - "numpy>=1.14"
-    - "pandas>=1.2.0"
-    - "bokeh>=1.4.0, <3.0.0"
+    - "python>=3.9, <3.12"
+    - "numpy>=1.20, <2.0"
+    - "pandas>=2.0, <3.0"
+    - "bokeh>=1.4, <3.0"
     - requests
     - numba
     - "paramtools>=0.18.0"

docs/usage/tcja_after_2025.md

@@ -26,7 +26,7 @@ expire).  Compatible input datasets include:
   [taxdata](https://github.com/PSLmodels/taxdata) repository and
   available only to those with access to the 2011 IRS/SOI PUF
 
-* several newer CSV-formatted input files created in the PSLmodels
+* the newer `tmd.csv` file generated in the PSLmodels
   [tax-microdata](https://github.com/PSLmodels/tax-microdata-benchmarking)
   repository that are based on the 2015 IRS/SOI PUF and on more recent CPS
   data, and are available only to those with access to the 2015

environment.yml

@@ -2,25 +2,23 @@ name: taxcalc-dev
 channels:
   - conda-forge
 dependencies:
-- "python<3.12"
-- curl
-- "numpy>=1.14"
-- "pandas>=1.2.0"
-- "bokeh>=1.4.0, <3.0.0"
+- "python>=3.9, <3.12"
+- "numpy>=1.20, <2.0"
+- "pandas>=2.0, <3.0"
+- "bokeh>=1.4, <3.0"
 - requests
-- aiohttp
 - numba
+- "paramtools>=0.18.0"
 - "fsspec<=0.8.7"
+- aiohttp
+- curl
 - pytest
-- pytest-pep8
 - pytest-xdist
 - pycodestyle
 - pylint
 - coverage
-- "paramtools>=0.18.0"
 - behresp
 - openpyxl
 - pip
 - pip:
     - jupyter-book
-    - pytest_harvest

pytest.ini

@@ -3,9 +3,14 @@ testpaths =
     taxcalc
 markers =
     requires_pufcsv
+    requires_tmdcsv
     pre_release
-    local
     compatible_data
+    local
     benefits
     itmded_vars
-    pep8
+    cpscsv_agg
+    pufcsv_agg
+    reforms
+    rtr
+    extend_tcja

taxcalc.egg-info/SOURCES.txt

@@ -24,6 +24,7 @@ taxcalc/puf_weights.csv.gz
 taxcalc/records.py
 taxcalc/records_variables.json
 taxcalc/taxcalcio.py
+taxcalc/tmd_weights.csv.gz
 taxcalc/utils.py
 taxcalc/utilsprvt.py
 taxcalc.egg-info/PKG-INFO

taxcalc/cli/tc.py

@@ -153,7 +153,11 @@ def cli_tc_main():
         sys.stderr.write(tcio.errmsg)
         sys.stderr.write('USAGE: tc --help\n')
         return 1
-    aging = inputfn.endswith('puf.csv') or inputfn.endswith('cps.csv')
+    aging = (
+        inputfn.endswith('puf.csv') or
+        inputfn.endswith('cps.csv') or
+        inputfn.endswith('tmd.csv')
+    )
     tcio.init(input_data=inputfn, tax_year=taxyear,
               baseline=args.baseline,
               reform=args.reform, assump=args.assump,

taxcalc/consumption.json

@@ -282,4 +282,4 @@
             }
         }
     }
-}
+}

taxcalc/data.py

@@ -276,6 +276,6 @@ def _read_weights(self, weights):
 
     def _extrapolate(self, year):
         """
-        Apply to dats variables the growth factor values for specified year.
+        Apply to data variables the growth factor values for specified year.
         """
         # Override this empty method in subclass

taxcalc/growdiff.json

@@ -500,4 +500,4 @@
             }
         }
     }
-}
+}

taxcalc/growdiff.py

@@ -7,6 +7,7 @@
 import os
 import numpy as np
 from taxcalc.parameters import Parameters
+from taxcalc.policy import Policy
 from taxcalc.growfactors import GrowFactors
 
 
@@ -26,8 +27,8 @@ class GrowDiff(Parameters):
     class instance: GrowDiff
     """
 
-    JSON_START_YEAR = 2013  # must be same as Policy.JSON_START_YEAR
-    DEFAULT_NUM_YEARS = 22  # must be same as Policy.DEFAULT_NUM_YEARS
+    JSON_START_YEAR = Policy.JSON_START_YEAR
+    DEFAULT_NUM_YEARS = Policy.DEFAULT_NUM_YEARS
     DEFAULTS_FILE_NAME = 'growdiff.json'
     DEFAULTS_FILE_PATH = os.path.abspath(os.path.dirname(__file__))
 
@@ -75,8 +76,13 @@ def apply_to(self, growfactors):
         """
         assert isinstance(growfactors, GrowFactors)
         for gfvn in GrowFactors.VALID_NAMES:
-            _gfvn = '_{}'.format(gfvn)
+            _gfvn = f'_{gfvn}'
             for i in range(0, self.num_years):
                 cyr = i + self.start_year
                 diff_array = getattr(self, _gfvn)
                 growfactors.update(gfvn, cyr, diff_array[i])
+
+    def set_rates(self):
+        """
+        Unimplemented base class method that is not used here.
+        """