City Research Online

Abstract

Non-random sample selection is a commonplace amongst many e mpirical studies and it appears when an output variable of interest is available only for a restricted non- random sub-sample of data. An extension of the generalized additive models for location, scale and shape which account s for non-random sample selection by introducing a selection equation is discussed. The proposed approach all ows for potentially any parametric distribution for the outcome variable, any parametric link function for the sele ction equation, several dependence structures between the (outcome and selection) equations through the use of copula e, and various types of covariate effects. Using a special case of the proposed model, it is shown how the score equation s are corrected for the bias deriving from non-random sample selection. Parameter estimation is carried out with in a penalized likelihood based framework. The empirical effectiveness of the approach is demonstrated through a sim ulation study and a case study. The models can be easily employed via the gjrm() function in the R package GJRM .

Publication Type:	Article
Additional Information:	© 2018 Elsevier B.V. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/
Publisher Keywords:	additive predictor, copula, marginal distribution, non-r andom sample selection, penalized regression spline, simultaneous equation estimation
Subjects:	H Social Sciences > HA Statistics Q Science > QA Mathematics
Departments:	Bayes Business School > Actuarial Science & Insurance
SWORD Depositor:	Symplectic Administrator

Export

Downloads