TY - JOUR
T1 - Large alphabet inference
AU - Painsky, Amichai
N1 - Publisher Copyright:
© The Author(s) 2023.
PY - 2023/12/1
Y1 - 2023/12/1
N2 - 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.
AB - 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.
KW - count data
KW - coverage probabilities
KW - large alphabet estimation
KW - multinomial proportions
KW - simultaneous inference
UR - http://www.scopus.com/inward/record.url?scp=85179806697&partnerID=8YFLogxK
U2 - 10.1093/imaiai/iaad049
DO - 10.1093/imaiai/iaad049
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
AN - SCOPUS:85179806697
SN - 2049-8772
VL - 12
JO - Information and Inference
JF - Information and Inference
IS - 4
M1 - iaad049
ER -