English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The sequential hypothesis testing problem is a class of statistical analyseswhere the sample size is not fixed in advance. Instead, the decision-processtakes in new observations sequentially to make real-time decisions for testingan alternative hypothesis against a null hypothesis until some stoppingcriterion is satisfied. In many common applications of sequential hypothesistesting, the data can be highly sensitive and may require privacy protection;for example, sequential hypothesis testing is used in clinical trials, wheredoctors sequentially collect data from patients and must determine when to stoprecruiting patients and whether the treatment is effective. The field ofdifferential privacy has been developed to offer data analysis tools withstrong privacy guarantees, and has been commonly applied to machine learningand statistical tasks.  In this work, we study the sequential hypothesis testing problem under aslight variant of differential privacy, known as Renyi differential privacy. Wepresent a new private algorithm based on Wald's Sequential Probability RatioTest (SPRT) that also gives strong theoretical privacy guarantees. We providetheoretical analysis on statistical performance measured by Type I and Type IIerror as well as the expected sample size. We also empirically validate ourtheoretical results on several synthetic databases, showing that our algorithmsalso perform well in practice. Unlike previous work in private hypothesistesting that focused only on the classical fixed sample setting, our results inthe sequential setting allow a conclusion to be reached much earlier, and thussaving the cost of collecting additional samples.

Private Sequential Hypothesis Testing for Statisticians: Privacy, Error Rates, and Sample Size