@CamDavidsonPilon I'm not quite sure I understand how you estimate the hazard rate through the cumulative hazard function. From what I understand, the cumulative hazard function can be estimated by taking the minus natural log of the Kaplan-Meier, or through the product summation of di/di. So I was wondering how you are estimating the hazard function from it without using a parametric distribution model? I can't find any clearer explanation of that from the documentation.
Good question, so the cumulative hazard rate is the integral of the hazard rate, so to recover the latter, one could use finite differences on the cumulative hazard, ex: h(t) = (H(t+d) - H(t))/d
However, this is quite a noisy estimate and only non-zero when H(t) changes over the interval [t, t+d]. So after differencing, one can apply a kernel smoother. https://en.wikipedia.org/wiki/Kernel_smoother
In general though, recovering the hazard rate without parametric models is difficult, hence why most people focus on the cumulative hazard.
@CamDavidsonPilon , thank you very much for your library. It is easy to use and very handy. I was the user who made this comment on stack overflow in response to which you opened CamDavidsonPilon/lifelines#438. What do you think about implementation?
:wave: a minor version 0.20.1 was just released, offers some nice performance improvements to models. https://github.com/CamDavidsonPilon/lifelines/releases/tag/v0.20.1
Hi there, I'm just getting started with lifelines and have hit a mental block. I'm trying to pass a duration and censor series to a KaplanMeierFitter; at first the durations had NaN values (which I thought would be fine because of the censor vector, but apparently not. Filtering those out now leaves me with: AttributeError: 'bool' object has no attribute 'any'. It's raised while checking for NaN and np.inf on the durations. I've confirmed there are no inf or nan values in the data I'm passing...
Hi @ningeo, let me try to help. Your duration array should be all numbers (no nans), and your
I suspect you are passing in a bool instead, ex:
event_observed variable should be an array of the same size as your duration array. Can you confirm that your event_observed is the same size?I suspect you are passing in a bool instead, ex:
kmf.fit(durations, True)
So, did you use the function
datetimes_to_durations from lifelines.utils to transform your data?
']
:wave: just released a minor version, 0.20.2 - support for left-censoring and qqplots. See change log here: https://github.com/CamDavidsonPilon/lifelines/releases/tag/v0.20.2
:wave: just released a minor version, 0.20.3, that adds some quality-of-life improvements for Kaplan-Meier users. https://github.com/CamDavidsonPilon/lifelines/releases
Maybe I should add this to the docs, though
I want to get the interval of predicted median value for Weibull model, and I write some codes to get it, but I'm not sure if this is corret, here is my code:
Thank you for your times!
class MyWeibullFitter(WeibullFitter):
@property
def median_confidence_interval_(self):
'''get the confidence interval of the median, must call after fit and plot'''
if self.median_ != np.inf:
self.timeline = np.linspace(self.median_, self.median_, 1)
return self.confidence_interval_survival_function_
else:
return NoneThank you for your times!
@githubhsss it's from a thesis, which is a pretty nice intro to a lot of common models: https://harvest.usask.ca/bitstream/handle/10388/etd-03302009-140638/JiezhiQiThesis.pdf
This actually isn't the most efficient way to compute the confidence intervals, but I think I'll expose a better way in the future
@CamDavidsonPilon I'am new to survival analysis, excuse me please if I'm wrong. I'm confused after reading wiki and papers about the confidence interval of survival function for parameter models, it would be a great help if you can give some references or documents about that!! Thanks a lot!
'''
@CamDavidsonPilon
Thanks for sharing the thesis again~
I'm dealing with some repeated events data(machine failure time data). Since a machine may have several failures and different machines have different number of failures, I think it's necessary to consider about repeated events and heterogeneity. Will frailty models help? Or any other advice? (^_^)/
Thanks for sharing the thesis again~
I'm dealing with some repeated events data(machine failure time data). Since a machine may have several failures and different machines have different number of failures, I think it's necessary to consider about repeated events and heterogeneity. Will frailty models help? Or any other advice? (^_^)/
@fuyb1992 ah yes, you may want to keep your
if self.median_ != np.inf check
@githubhsss frailty, is one solution, though it's not in lifelines (but is in R's survival). Another option is to use
cluster_col is CoxPHFitter: https://lifelines.readthedocs.io/en/latest/Examples.html#correlations-between-subjects-in-a-cox-model. Another solution is to strata-ify per machine in the CoxPHFitter.
Thanks a lot! I'm trying to understand the confidence interval of survival function for parameter models, the Taylor expansions method is mentioned a lot , and the Jacobian-vector product is used in lifelines code. I'm confused with the relationship between them, it would be a great help if you could give some references or documents about the implementation method. Thank you for your time!!