A new link function for the prediction of binary variables
Gheno Gloria ()
Additional contact information
Gheno Gloria: Free University of Bolzano-Bozen, Italy
Croatian Review of Economic, Business and Social Statistics, 2018, vol. 4, issue 2, 67-77
If there are no heavy sanctions in place to prevent it, the problem of the cancellation of appointments can lead to huge economic losses and can have a significant impact on underutilized resources of healthcare facilities. A good model to predict the appointment cancellations could be an effective solution to this problem. Therefore, a new Bayesian method is proposed to estimate accurately the probability of the cancellation of visits to healthcare institutions based on specific factors such as age. This model uses the regression for binary variables, linking the explanatory variables to the probability of appearance at a previously made appointment with a new weighted function and estimating the parameters with the Bayesian method. The goodness of the new method is demonstrated by applying it to a real case and by comparing it to other methodologies. Therefore, the advantages of the proposed method are exposed and possible real-world applications are described.
Keywords: Bayesian method; binary variables; cancellation prediction; heath care; link function (search for similar items in EconPapers)
JEL-codes: C11 C50 C51 I11 (search for similar items in EconPapers)
References: Add references at CitEc
Citations Track citations by RSS feed
Downloads: (external link)
https://www.degruyter.com/view/j/crebss.2018.4.iss ... -0014.xml?format=INT (text/html)
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
Persistent link: https://EconPapers.repec.org/RePEc:vrs:crebss:v:4:y:2018:i:2:p:67-77:n:8
Access Statistics for this article
Croatian Review of Economic, Business and Social Statistics is currently edited by Dragan Bagić, Ksenija Dumičić and Nataša Erjavec
More articles in Croatian Review of Economic, Business and Social Statistics from Sciendo
Bibliographic data for series maintained by Peter Golla ().