Shan Huang
Shan Huang
Just a remark for the future work (not part of this pull request): - for a normal distribution, the returned effect size is already normalized - for a Poisson distribution,...
...using pre-experiment data. See this post: https://booking.ai/how-booking-com-increases-the-power-of-online-experiments-with-cuped-995d186fff1d The underlying method is [[Deng, Xu, Kohavi, & Walker, 2013]](http://www.exp-platform.com/Documents/2013-02-CUPED-ImprovingSensitivityOfControlledExperiments.pdf).
Because of multiple testing correction, confidence interval will be broader. Besides, we might need to consider adding p-value into the result. This is a reminder for us to discuss the...
Here in this line we have specified some additional parameters to StanModel.sampling: https://github.com/zalando/expan/blob/dev/expan/core/early_stopping.py#L189 Do we know what are the effects of specifying ```stepsize``` and ```adapt_delta``` in this case?
The following code in `Experiment.delta` ``` pattern = '([a-zA-Z][0-9a-zA-Z_]*)' self.kpis.loc[:,kpiName] = eval(re.sub(pattern, r'self.kpis.\1.astype(float)', dk['formula'])) ``` will only work for fomulas with simple ratios (and also too hacky). However, I found...
Auto-reviewers: @NiharikaRay @matthewwardrop @earthmancash @danfrankj To whom it may concern, thank you for releasing this as open source project, it is just great! I have a question (instead of an...