fit_novelty: Fit Model for detecting novelty effect from Averate Treatment...

In okiyuki99/abstats: abstats

Description

Fit Model for detecting novelty effect from Averate Treatment Effect

Usage

fit_novelty(ate, alpha = 0.35, gamma = 2)

Arguments

ate	Averate Treatment Effect, i.e. Expecttaion of Treatment - Expecttaion of Control
alpha	parameter alpha
gamma	parameter gamma

Value

linear model object

See Also

How A/B Tests Could Go Wrong: Automatic Diagnosis of Invalid Online Experiments, WSDM2019

Examples

## Not run: 
ate <- rev(1.3^(rnorm(14, 0, 0.3) + 1:14) / 10)
fit <- fit_novelty(ate)
summary(fit)
fit <- fit_novelty(ate, alpha = 0.3, gamma = 2)
summary(fit)

## End(Not run)

okiyuki99/abstats documentation built on Jan. 29, 2020, 6:48 p.m.