https://topazape.dev/ Statistical modeling and causal inference, with applications in advertising effectiveness measurement. quarto-1.10.18 Wed, 29 Jul 2026 00:00:00 GMT https://topazape.dev/posts/bls-fieller-ci/ previous post showed that the log-transform delta method CI for achieves its nominal 95% coverage when the expected number of positive control responses is 20 or more. Below that, the delta method becomes conservative: the term in the SE formula inflates the interval beyond what the stated confidence level implies. The bootstrap percentile CI, which does not depend on the Taylor approximation, maintained near-nominal coverage across the full range. ]]> brand-lift statistics https://topazape.dev/posts/bls-fieller-ci/ Wed, 29 Jul 2026 00:00:00 GMT https://topazape.dev/posts/bls-bootstrap-validation/ previous post derived a log-transform delta method CI for the risk ratio : ]]> brand-lift statistics https://topazape.dev/posts/bls-bootstrap-validation/ Wed, 15 Jul 2026 00:00:00 GMT https://topazape.dev/posts/bls-delta-method-rr/ , where and are the positive response rates in the exposed and control groups, is the most direct measure of how much exposure shifted the outcome. An of means the exposed group responded positively 12% more often than control. Platforms typically report as “relative lift”, but the underlying quantity is a ratio of two binomial proportions. ]]> brand-lift statistics https://topazape.dev/posts/bls-delta-method-rr/ Wed, 01 Jul 2026 00:00:00 GMT