P-splines and GAMLSS: a powerful combination,with an application to zero-adjusted distributions

Abstract

P-splines are a versatile statistical modelling tool, dealing with nonlinear relationships between the response and explanatory variable(s). GAMLSS is a distributional regression framework which allows modelling of a response variable using any parametric distribution. The combination of the two methodologies provides one of the most powerful tools in modern regression analysis. This article discusses the application of the two techniques when the response variable is zero-adjusted (or semi-continuous), which combines a point probability at zero with a positive continuous distribution.

Keywords

P-splines gamlss Box-Cox t distribution zero-adjusted distribution zero-heavy distribution semi-continuous distribution

Get full access to this article

View all access options for this article.

References

Aitkin

(1987) Modelling variance heterogeneity in normal regression using GLIM. Applied Statistics , 36, 332–339.

Aitkin

(2018) A history of the GLIM statistical package. International Statistical Review , 86, 275–299.

Belasco

and Ghosh

(2012) Modelling semi-continuous data using mixture regression models with an application to cattle production yields. The Journal of Agricultural Science , 150, 109–121.

De Bastiani

, Stasinopoulos

, Rigby

, Heller

and Silva

(2022) Bucket plot: A visual tool for skewness and kurtosis comparisons. Brazilian Journal of Probability and Statistics , 36, 421–440.

Eilers

PHC

and Marx

(1996) Flexible smoothing with B-splines and penalties. Statistical Science , 11, 89–102.

Eilers

PHC

and Marx

(2010) Splines, knots and penalties. Wiley Interdisciplinary Reviews: Computational Statistics , 2, 637–653.

Eilers

PHC

and Marx

(2021) Practical smoothing. The Joys of P-splines . Cambridge University Press.

Hastie

and Tibshirani

(1990) Generalized additive models . Chapman and Hall.

Heller

, Stasinopoulos

and Rigby

(2006) The zero-adjusted inverse Gaussian distribution as a model for insurance data. In: Proceedings of the 21st International Workshop on Statistical Modelling, Galway, Ireland , pages 226–233.

10.

Houghton

, Nowlin

and Walker

(2019) From fantasy to reality: The role of fantasy sports in sports betting and online gambling. Journal of Public Policy and Marketing , 38, 332–353.

11.

Kauermann

(2010). Penalized splines, mixed models and Bayesian ideas. In: Statistical modelling and regression structures , edited by Kneib

and Tutz

. Physica-Verlag HD.

12.

Kneib

(2013) Beyond mean regression. Statistical Modelling , 13, 275–3036.

13.

Liu

, Shih

YCT

, Strawderman

, Zhang

, Johnson

and Chai

(2019) Statistical analysis of zero-inflated nonnegative continuous data: a review. Statistical Science , 34, 253–279.

14.

Neelon

, O’Malley

and Smith

(2016a) Modeling zero-modified count and semicontinuous data in health services research Part 1: background and overview. Statistics in Medicine , 35, 5070–5093.

15.

Neelon

, O’Malley

and Smith

(2016b) Modeling zero-modified count and semicontinuous data in health services research part 2: case studies. Statistics in Medicine , 35, 5094–5112.

16.

Nelder

and Wedderburn

RWM

. (1972). Generalized linear models. Journal of the Royal Statistical Society, Series A , 135, 370–384.

17.

O’Sullivan

(1986) A statistical perspective on ill-posed inverse problems (with discussion). Statistical Science , 1, 505–527.

18.

Popuri

, Neerchal

and Mehta

(2015) Comparison of Linear and Tobit Modeling of Downscaled Daily Precipitation over the Missouri River Basin Using MIROC5. In: Machine learning and data mining approaches to climate science , edited by Lakshmanan

, Gilleland

, McGovern

and Tingley

. Springer.

19.

Rigby

and Stasinopoulos

(2005) Generalized additive models for location, scale and shape. Applied Statistics, Series C , 54, 507–554.

20.

Rigby

and Stasinopoulos

(2006) Using the Box-Cox t distribution in GAMLSS to model skewness and kurtosis. Statistical Modelling , 6, 209–229.

21.

Rigby

and Stasinopoulos

(2012) Automatic smoothing parameter selection in GAMLSS with an application to centile estimation. Statistical Methods in Medical Research , 23, 318–332.

22.

Rigby

, Stasinopoulos

, Heller

and De Bastiani

(2019) Distributions for modeling location, scale, and shape: Using GAMLSS in R . Chapman & Hall/CRC, Boca Raton.

23.

Rodríguez-Álvarez

and Oviedo de la Fuente

(2021) SOP: Generalised additive P-spline regression models estimation . R package version 1.0.

24.

Smyth

(1989) Generalized linear models with varying dispersion. Journal of the Royal Statistical Society, Series B , 51, 47–60.

25.

Stasinopoulos

and Rigby

(2008) Generalized additive models for location scale and shape (GAMLSS) in R. Journal of Statistical Software , 23, 1–46.

26.

Stasinopoulos

, Rigby

, Heller

, Voudouris

and De Bastiani

(2017) Flexible regression and smoothing: using GAMLSS in R . Chapman & Hall/CRC, Boca Raton.

27.

Tobin

(1958) Estimation of relationships for limited dependent variables. Econometrica: Journal of the Econometric Society , 26, 24–36.

28.

Tweedie

MCK

(1984) An index which distinguishes between some important exponential families. In: Statistics: Applications and New Directions: Proceedings of Indian Statistical Institute Golden Jubilee International Conference , edited by Ghosh

and Roy

, pages 579–604.

29.

Umlauf

, Klein

, Simon

and Zeileis

(2021) BAMLSS: A lego toolbox for flexible Bayesian regression (and beyond). Journal of Statistical Software , 100, 1–53. Url https://10.18637/jss.v100.i04

30.

Van Buuren

and Fredriks

(2001) Worm plot: A simple diagnostic device for modelling growth reference curves. Statistics in Medicine , 20, 1259–1277.

31.

Van Ogtrop

, Vervoort

, Heller

, Stasinopoulos

and Rigby

(2011) Long-range forecasting of intermittent streamflow. Hydrology and Earth System Sciences , 15, 3343–3354.

32.

Wood

and Fasiolo

(2017) A generalized Fellner-Schall method for smoothing parameter optimization with application to Tweedie location, scale and shape models. Biometrics , 73, 1071–1081.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.18 MB

0.00 MB