A Latent Class Panel Tobit Framework:Application to Modelling Charitable Donations

We extend the latent class literature by re-examining censored variable analysis within a panel data context. Specifically, we extend the standard latent class tobit panel approach to simultaneously include random effects, to allow for heteroskedasticity and to incorporate the inverse hyperbolic s...

Full description

Bibliographic Details
Main Authors: Greene, William, Brown, S., Harris, Mark N., Taylor, K.
Format: Journal Article
Published: Wiley-Blackwell Publishing 2013
Subjects:
Online Access:http://webmeets.com/files/papers/res/2014/107/LC%20Charity%20July%202013.pdf
http://hdl.handle.net/20.500.11937/15020
_version_ 1848748781319749632
author Greene, William
Brown, S.
Harris, Mark N.
Taylor, K.
Taylor, K.
author_facet Greene, William
Brown, S.
Harris, Mark N.
Taylor, K.
Taylor, K.
author_sort Greene, William
building Curtin Institutional Repository
collection Online Access
description We extend the latent class literature by re-examining censored variable analysis within a panel data context. Specifically, we extend the standard latent class tobit panel approach to simultaneously include random effects, to allow for heteroskedasticity and to incorporate the inverse hyperbolic sine (IHS) transformation of the dependent variable. The IHS transformation ensures robustness to nonnormality in the original (untransformed) dependent variable. We then apply this framework to modelling charitable donations, an interesting application given the potential for divergent groups of individuals in the population with regard to their donating behaviour, which we uncover by the latent class approach. Our findings, which are based on U.S. panel data drawn from five waves of the Panel Study of Income Dynamics, do suggest two distinct classes. There is a clear disparity between the probabilities of zero donations across these classes, with one class dominated by the observed zero givers and associated with relatively low levels of predicted giving. We find clear evidence of both heteroskedasticity and random effects. All IHS parameters were significantly different from zero and different across classes. In combination, these findings endorse the importance of our three modelling extensions.
first_indexed 2025-11-14T07:10:30Z
format Journal Article
id curtin-20.500.11937-15020
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T07:10:30Z
publishDate 2013
publisher Wiley-Blackwell Publishing
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-150202017-01-30T11:47:18Z A Latent Class Panel Tobit Framework:Application to Modelling Charitable Donations Greene, William Brown, S. Harris, Mark N. Taylor, K. Taylor, K. Tobit Panel Data Latent Class Donations Charity We extend the latent class literature by re-examining censored variable analysis within a panel data context. Specifically, we extend the standard latent class tobit panel approach to simultaneously include random effects, to allow for heteroskedasticity and to incorporate the inverse hyperbolic sine (IHS) transformation of the dependent variable. The IHS transformation ensures robustness to nonnormality in the original (untransformed) dependent variable. We then apply this framework to modelling charitable donations, an interesting application given the potential for divergent groups of individuals in the population with regard to their donating behaviour, which we uncover by the latent class approach. Our findings, which are based on U.S. panel data drawn from five waves of the Panel Study of Income Dynamics, do suggest two distinct classes. There is a clear disparity between the probabilities of zero donations across these classes, with one class dominated by the observed zero givers and associated with relatively low levels of predicted giving. We find clear evidence of both heteroskedasticity and random effects. All IHS parameters were significantly different from zero and different across classes. In combination, these findings endorse the importance of our three modelling extensions. 2013 Journal Article http://hdl.handle.net/20.500.11937/15020 http://webmeets.com/files/papers/res/2014/107/LC%20Charity%20July%202013.pdf Wiley-Blackwell Publishing restricted
spellingShingle Tobit
Panel Data
Latent Class
Donations
Charity
Greene, William
Brown, S.
Harris, Mark N.
Taylor, K.
Taylor, K.
A Latent Class Panel Tobit Framework:Application to Modelling Charitable Donations
title A Latent Class Panel Tobit Framework:Application to Modelling Charitable Donations
title_full A Latent Class Panel Tobit Framework:Application to Modelling Charitable Donations
title_fullStr A Latent Class Panel Tobit Framework:Application to Modelling Charitable Donations
title_full_unstemmed A Latent Class Panel Tobit Framework:Application to Modelling Charitable Donations
title_short A Latent Class Panel Tobit Framework:Application to Modelling Charitable Donations
title_sort latent class panel tobit framework:application to modelling charitable donations
topic Tobit
Panel Data
Latent Class
Donations
Charity
url http://webmeets.com/files/papers/res/2014/107/LC%20Charity%20July%202013.pdf
http://hdl.handle.net/20.500.11937/15020