Large alphabet inference

Amichai Painsky*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


Consider a finite sample from an unknown multinomial distribution. Inferring the underlying multinomial parameters is a basic problem in statistics and related fields. Currently known methods focus on classical regimes where the sample is large, or both the sample and the alphabet are small. In this work we study the complementary large alphabet regime, as we consider the case where the number of samples is comparable with (or even smaller than) the alphabet size. We introduce a novel inference scheme that significantly improves upon currently known methods. Our proposed scheme is robust, easy to apply and provides favourable performance guarantees.

Original languageEnglish
Article numberiaad049
JournalInformation and Inference
Issue number4
StatePublished - 1 Dec 2023


FundersFunder number
Israel Science Foundation963/21
Israel Science Foundation


    • count data
    • coverage probabilities
    • large alphabet estimation
    • multinomial proportions
    • simultaneous inference


    Dive into the research topics of 'Large alphabet inference'. Together they form a unique fingerprint.

    Cite this