Input/Output

Compute the log-rank P-value and survival curves based on kaplan meier.

param time_event

Numpy array with survival-time in years/months/days (not a datetime!)

type time_event

Float.

param censoring

numpy array with censoring (1=event, 0=ongoing). At the time you want to make inferences about durations, it is possible, likely true, that not all the death events have occured yet. In case of patients, you would like to put 1=death as an event.

type censoring

array-like

param labx

Class labels. Each class label is seperately plotted.

type labx

array-like of type string or int

param verbose

Verbosity messages.

type verbose

int, (default: 3)

returns

• logrank_P (float) : P-value

• logrank_Z (float) : Z-score

• logrank (float) : fitted logrank_test model

• labx (list) : Class labels

• uilabx (list) : Unique Class labels

• time_event (float) : Time to event

• censoring (bool) : Censored or not

rtype

dict()

Examples

>>> # Import library
>>> import kaplanmeier as km
>>>
>>> # Example data
>>> df = km.example_data()
>>>
>>> # Fit
>>> results = km.fit(df['time'], df['Died'], df['group'])
>>>
>>> # Plot
>>> km.plot(results)
>>>
>>> km.plot(results, cmap='Set1', cii_lines=True, cii_alpha=0.05)
>>> km.plot(results, cmap=[(1, 0, 0),(0, 0, 1)])
>>> km.plot(results, cmap='Set1', methodtype='custom')
>>>
>>> results['logrank_P']
>>> results['logrank_Z']

---

---