TY - CONF

T1 - Modelling a proportion response variable using generalised additive models for location scale and shape

AU - Enea, Marco

PY - 2015

Y1 - 2015

N2 - In this paper two alternative approaches are proposed to model aresponse variable Y measured on the interval from zero to one, including both zeroand one. The first proposed model employs a flexible four parameter distributionfor 0 < Y < 1, for example a logit skew exponential power distribution, inflatedby including point probabilities at 0 and 1. The second proposed model is ageneralised Tobit model, obtained from a flexible four parameter distributionon (-infinity;+infinity), for example the skew exponential power distribution, by censoringbelow 0 and above 1. The proposed models are applied to a real data set andcompared with current popular models.

AB - In this paper two alternative approaches are proposed to model aresponse variable Y measured on the interval from zero to one, including both zeroand one. The first proposed model employs a flexible four parameter distributionfor 0 < Y < 1, for example a logit skew exponential power distribution, inflatedby including point probabilities at 0 and 1. The second proposed model is ageneralised Tobit model, obtained from a flexible four parameter distributionon (-infinity;+infinity), for example the skew exponential power distribution, by censoringbelow 0 and above 1. The proposed models are applied to a real data set andcompared with current popular models.

KW - GAMLSS

KW - Generalised Tobit model

KW - Logit skew exponential power
distribution.

KW - GAMLSS

KW - Generalised Tobit model

KW - Logit skew exponential power
distribution.

UR - http://hdl.handle.net/10447/247272

M3 - Other

ER -