Some approximations of the logistic distribution with application to the covariance matrix of logistic regression
2014 (English)In: Statistics and Probability Letters, ISSN 0167-7152, Vol. 85, 63-68 p.Article in journal (Refereed) Published
In this paper, we show that a two-component normal mixture model provides a good approximation to the logistic distribution. This approximation is an improvement over using the normal distribution and is comparable to using the t-distribution as approximating distributions. The results from using the mixture model is exemplified by finding an approximative analytic expression for the covariance matrix of logistic regression using normally distributed random regressors.
Place, publisher, year, edition, pages
2014. Vol. 85, 63-68 p.
Density, Gaussian, Mixture, Normal, t -distribution
Probability Theory and Statistics
Research subject Statistics
IdentifiersURN: urn:nbn:se:uu:diva-221566DOI: 10.1016/j.spl.2013.11.007ISI: 000331665000009OAI: oai:DiVA.org:uu-221566DiVA: diva2:709472