Knowledge Discovery, Metrics and Uncertainty

Search:

DAY

WEEK

MONTH

YEAR

Select your country:
Sign up | Log in

Knowledge Discovery

Metrics

Uncertainty

Variance and significance in large-scale online services

The Unofficial Google Data Science Blog

JANUARY 14, 2016

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. We must therefore maintain statistical rigor in quantifying experimental uncertainty.

Experimentation

Experimentation Statistics Metrics Measurement

LSOS experiments: how I learned to stop worrying and love the variability

The Unofficial Google Data Science Blog

FEBRUARY 29, 2016

The result is that experimenters can’t afford to be sloppy about quantifying uncertainty. These typically result in smaller estimation uncertainty and tighter interval estimates. Variance reduction through conditioning Suppose, as an LSOS experimenter, you find that your key metric varies a lot by country and time of day.

Experimentation

Experimentation Statistics Metrics Measurement

Webinars

How to Achieve High-Accuracy Results When Using LLMs

MORE WEBINARS

Data Leaders Brief

Variance and significance in large-scale online services

LSOS experiments: how I learned to stop worrying and love the variability

Webinars

Stay Connected