A Bayesian approach for combining probability and non-probability samples surveys

Camilla Salvatore, Silvia Biffignandi, Joseph Sakshaug, Bella Struminskaya, Arkadiusz Wiśniowski

Research output: Chapter in Book/Conference proceedingConference contributionpeer-review

103 Downloads (Pure)

Abstract

Our paper proposes a method of combining probability and non-probability samples to improve analytic inference on logistic regression model parameters. A Bayesian framework is considered where only a small probability sample is available and the information from a parallel non-probability sample is provided naturally through the prior. A simulation study is run applying several informative priors. Comparisons on the performance of the models are studied with reference to their mean-squared error (MSE). In general, the informative priors reduce the MSE or, in the worst-case scenario, perform equivalently to non-informative priors.
Original languageEnglish
Title of host publicationBook of Short Papers
Subtitle of host publicationSIS 2022 51st Scientific Meeting of the Italian Statistical Society, Caserta, 22-24 June
EditorsAntonio Balzanella, Matilde Bini, Carlo Cavicchia, Rosanna Verde
Place of PublicationMilan
PublisherPearson Italia
Pages717-722
Number of pages6
ISBN (Electronic)9788891932310
Publication statusPublished - Jun 2022
Event51st Scientific Meeting of the Italian Statistical Society - Caserta, Italy
Duration: 22 Jun 202224 Jun 2022

Conference

Conference51st Scientific Meeting of the Italian Statistical Society
Abbreviated titleSIS 2022
Country/TerritoryItaly
CityCaserta
Period22/06/2224/06/22

Keywords

  • selection bias
  • data integration
  • Bayesian inference

Fingerprint

Dive into the research topics of 'A Bayesian approach for combining probability and non-probability samples surveys'. Together they form a unique fingerprint.

Cite this