binsmooth 0.13

binsmooth - Better Estimates from Binned Income Data.

binsmooth

Python implementation of "Better Estimates from Binned Income Data"

Better Estimates from Binned Income Data: Interpolated CDFs and Mean-Matching
Paul T. von Hippel, David J. Hunter, McKalie Drown
Sociological Science
Volume 4, Number 26, Pages 641-655
2017

Originally implemented in the R package binsmooth.

Usage

from binsmooth import BinSmooth

bin_edges = np.array([0, 18200, 37000, 87000, 180000])
counts = np.array([0, 7527, 13797, 75481, 50646, 803])

bs = BinSmooth()
bs.fit(bin_edges, counts)

# Print median estimate
print(bs.inv_cdf(0.5))

Installation

Install via pip

pip install binsmooth

Improvements

Better tail estimate by bounded optimisation rather than the adhoc search method found in the R implementation.

More precise inverse CDF by dynamically sampling the CDF. This is done by sampling proportional to the steepness of the CDF i.e. sampling more in areas where the CDF is steeper.

Warnings

Results do not exactly match R binsmooth because:

1. we take a different approach to estimating the tail (upper bound)
2. choice of spline interpolation

This implementation uses scipy's PchipInterpolator which implements [1], while the default interpolator in the R implementation is [2]. The interpolator in the R implementation can be changed to [1] by setting monoMethod="monoH.FC".

Accuracy is dependent on the mean of the distribution. If you do not supply a mean, then one will be estimated in an adhoc manner and the accuracy of estimates may be poor.

References

[1]: Fritsch, F. N. and Carlson, R. E. (1980). Monotone piecewise cubic interpolation. SIAM Journal on Numerical Analysis
[2]: Hyman, J. M. (1983). Accurate monotonicity preserving cubic interpolation. SIAM Journal on Scientific and Statistical Computing