Fix Age Variables

[andersonfrailey]

This PR addresses concerns raised by @MaxGhenis in issue #164. I've corrected the nu05, nu13, and n24 variables in the CPS file, as well as modified the scripts to create the PUF. Interestingly, I got the same MD5 output from the PUF after rerunning the scripts with the included modification.

cc @martinholmer

[MaxGhenis]

Interestingly, I got the same MD5 output from the PUF after rerunning the scripts with the included modification.

Are the source fields like ICPS03 ints?

[andersonfrailey]

@MaxGhenis, no. We've written the Python scripts so that those variables are incremented while dependents are added to the tax unit, rather than after the fact.

[martinholmer]

@andersonfrailey, using the new puf.csv file generated by the changes in #172, will the following Tax-Calculator test pass (after removing the xfail decorator)? If not, why does the test fail? Is the test logic incorrect? Or are there still problems with the age count variables in the puf.csv file?

@pytest.mark.requires_pufcsv
@pytest.mark.xfail
def test_ubi_n_variables(puf_path):
    """
    Ensure that the three UBI n* variables add up to XTOT variable,
    recognizing that XTOT values are often capped in the IRS-SOI PUF,
    so that XTOT < NSUM might not indicate any data inconsistency.
    """
    pufdf = pd.read_csv(puf_path)
    xtot = pufdf['XTOT']
    nsum = pufdf['nu18'] + pufdf['n1820'] + pufdf['n21']
    if not np.sum(xtot > nsum) == 0:
        print('number xtot > nsum is:', np.sum(xtot > nsum))
        assert 'XTOT' <= '(nu18+n1820+n21)'

[andersonfrailey]

@martinholmer, that will still fail. I haven't been able to address that issue just yet.

[andersonfrailey]

I've updated this PR to include the last few PR's that have been merged. As I mentioned yesterday, I will be offline until the 10th, but this PR is ready for review and can be merged by one of the other maintainers if there are no issues with it while I'm gone.