Remove Knowledge Discovery Remove Reference Remove Statistics Remove Testing
article thumbnail

Changing assignment weights with time-based confounders

The Unofficial Google Data Science Blog

This post considers a common design for an OCE where a user may be randomly assigned an arm on their first visit during the experiment, with assignment weights referring to the proportion that are randomly assigned to each arm. For example, imagine a fantasy football site is considering displaying advanced player statistics.

article thumbnail

Performing Non-Compartmental Analysis with Julia and Pumas AI

Domino Data Lab

Once all packages have been imported, we can move on to loading our test data. We can group by study arm and calculate various statistics as mean and standard deviation. We can then proceed with pharmacokinetic modeling, testing the goodness of fit of various models. References. [1] Non Compartmental Analysis.

Metrics 59
article thumbnail

Variance and significance in large-scale online services

The Unofficial Google Data Science Blog

Unlike experimentation in some other areas, LSOS experiments present a surprising challenge to statisticians — even though we operate in the realm of “big data”, the statistical uncertainty in our experiments can be substantial. Because individual observations have so little information, statistical significance remains important to assess.