CoCalc -- nsfg2.py

Think Stats by Allen B. Downey Think Stats is an introduction to Probability and Statistics for Python programmers.

This is the accompanying code for this book.

Website: http://greenteapress.com/wp/think-stats-2e/

Views: ⁷¹¹⁹
License: GPL3

1
"""This file contains code used in "Think Stats",
2
by Allen B. Downey, available from greenteapress.com
3

4
Copyright 2014 Allen B. Downey
5
License: GNU GPLv3 http://www.gnu.org/licenses/gpl.html
6
"""
7

8
from __future__ import print_function
9

10
import numpy as np
11

12
import thinkstats2
13

14
def MakeFrames():
15
    """Reads pregnancy data and partitions first babies and others.
16

17
    returns: DataFrames (all live births, first babies, others)
18
    """
19
    preg = ReadFemPreg()
20

21
    live = preg[preg.outcome == 1]
22
    firsts = live[live.birthord == 1]
23
    others = live[live.birthord != 1]
24

25
    assert(len(live) == 14292)
26
    assert(len(firsts) == 6683)
27
    assert(len(others) == 7609)
28

29
    return live, firsts, others
30

31

32
def ReadFemPreg(dct_file='2006_2010_FemPregSetup.dct',
33
                dat_file='2006_2010_FemPreg.dat.gz'):
34
    """Reads the NSFG 2006-2010 pregnancy data.
35

36
    dct_file: string file name
37
    dat_file: string file name
38

39
    returns: DataFrame
40
    """
41
    dct = thinkstats2.ReadStataDct(dct_file, encoding='iso-8859-1')
42
    df = dct.ReadFixedWidth(dat_file, compression='gzip')
43
    CleanFemPreg(df)
44
    return df
45

46

47
def CleanFemPreg(df):
48
    """Recodes variables from the pregnancy frame.
49

50
    df: DataFrame
51
    """
52
    # mother's age is encoded in centiyears; convert to years
53
    df.agepreg /= 100.0
54

55
    # birthwgt_lb contains at least one bogus value (51 lbs)
56
    # replace with NaN
57
    df.loc[df.birthwgt_lb1 > 20, 'birthwgt_lb1'] = np.nan
58
    
59
    # replace 'not ascertained', 'refused', 'don't know' with NaN
60
    na_vals = [97, 98, 99]
61
    df.birthwgt_lb1.replace(na_vals, np.nan, inplace=True)
62
    df.birthwgt_oz1.replace(na_vals, np.nan, inplace=True)
63

64
    # birthweight is stored in two columns, lbs and oz.
65
    # convert to a single column in lb
66
    # NOTE: creating a new column requires dictionary syntax,
67
    # not attribute assignment (like df.totalwgt_lb)
68
    df['totalwgt_lb'] = df.birthwgt_lb1 + df.birthwgt_oz1 / 16.0    
69

70
    # due to a bug in ReadStataDct, the last variable gets clipped;
71
    # so for now set it to NaN
72
    df.phase = np.nan
73

74

75
def main():
76
    live, firsts, others = MakeFrames()
77

78

79
if __name__ == '__main__':
80
    main()
81

82

83

84