def test_multivariate_logrank_on_dd_dataset():
"""
library('survival')
dd = read.csv('~/code/lifelines/lifelines/datasets/dd.csv')
results = survdiff(Surv(duration, observed)~regime, data=dd, rho=0)
results[5]
"""
dd = load_dd()
results = stats.multivariate_logrank_test(dd["duration"], dd["regime"], dd["observed"])
assert abs(results.test_statistic - 322.5991) < 0.0001
from lifelines.datasets import load_dd
from lifelines import KaplanMeierFitter
from matplotlib import pyplot as plt
data = load_dd()
from lifelines.datasets import load_rossi
from lifelines import CoxPHFitter
from infra import *
import pandas
FILTER_IN = "filter in"
FILTER_OUT = "filter out"
OS_FIELDS = ["_OS", "_OS_IND"]
pheno_start = 1
pheno_limit = 135
def cox_phenotype(test_independently, filtered_out, filtered_in, filter_type,
filter_na_by_rows, phenotype_file_name, survival_file_name):
phenotype_dataset = load_phenotype_data(
phenotype_file_name=phenotype_file_name, phenotype_list_path=None)
survival_dataset = load_survival_data(survival_file_name,
survival_list_path=None)
pheno_survival_integrated = {}
for cur_pheno in phenotype_dataset[1:]:
pheno_survival_integrated[
cur_pheno[0]] = cur_pheno[pheno_start:pheno_limit]
for cur_survival in survival_dataset[1:]:
if pheno_survival_integrated.has_key(cur_survival[0]):
pheno_survival_integrated[
cur_survival[0]] = pheno_survival_integrated[