surpyval 0.4.0

A python package for survival analysis

SurPyval - Survival Analysis in Python

Yet another Python survival analysis tool. This is another pure python survival analysis tool so why was it needed? The intent of this package was to closely mimic the scipy API as close as possible, other survival analysis packages don't completely mimic that API. Further, there was a gap in the other tools, specifically, arbitrary combinations of observed, censored, and truncated data. The current capability is as follows:

Method	Para/Non-Para	Observed	Censored	Truncated
MLE	Parametric	Yes	Yes	Yes
MPP	Parametric	Yes	Yes	Limited
MSE	Parametric	Yes	Yes	Limited
MOM	Parametric	Yes	No	No
MPS	Parametric	Yes	Yes	No
Kaplan-Meier	Non-Parametric	Yes	Right only	Left only
Nelson-Aalen	Non-Parametric	Yes	Right only	Left only
Fleming-Harrington	Non-Parametric	Yes	Right only	Left only
Turnbull	Non-Parametric	Yes	Yes	Yes

SurPyval also offers many different distributions for users, and because of the flexible implementation adding new distributions is easy. The available distributions are:

Distribution
Weibull
Normal
LogNormal
Gamma
Beta
Uniform
Exponential
Exponentiated Weibull
Gumbel
Logistic
LogLogistic

This project spawned from a Reliaility Engineering project, as such there is a focus on visualising distribution fits with probability plots. This is combined with extremely flexible non-parametric fitting estimates. So each parametric.

Install and Quick Intro

SurPyval can be installed via pip using the PyPI repository

pip install surpyval

If you're familiar with survival analysis, and Weibull plotting, the following is a quick start.

from surpyval import Weibull
from surpyval.datasets import BoforsSteel

# Fetch some data that comes with SurPyval
data = BoforsSteel.df

x = data['x']
n = data['n']

model = Weibull.fit(x=x, n=n, offset=True)
model.plot();

Documentation

SurPyval is well documented, and improving, at the main documentation.

Contact

Email derryn if you want any features or to see how SurPyval can be used for you.