Abstract
This article introduces regression quantile models using both RS (Ramberg and Schmeiser, Commun Assoc Comput Mach 17:78–82, 1974) and FKML (Freimer et al., Commun Stat 17(10):3547–3567, 1988) generalised lambda distributions (GLD) and demonstrates the versatility of proposed models for a range of linear/non-linear and heteroscedastic/homoscedastic empirical data. Owing to the rich shapes of GLDs, GLD quantile regression is a competitive flexible model compared to standard quantile regression. The proposed method has some major advantages: (1) it provides a reference line which is very robust to outliers with the attractive property of zero mean residuals and (2) it gives a unified, elegant quantile regression model from the reference line with smooth regression coefficients across different quantiles. The proposed method has wide applications given the flexibility of GLDs. The goodness of fit of the proposed model can be assessed via QQ plots and the Kolmogorov–Smirnov test, to ensure the appropriateness of the statistical inference under consideration.
Similar content being viewed by others
References
Cheng, R.C.H., Amin, N.A.K.: Estimating Parameters in Continuous Univariate Distributions with a Shifted Origin. J. R. Stat. Soc. B 45(3), 394–403 (1983)
Cheng, R.C.H., Stephens, M.A.: A goodness-of-fit test using Moran’s statistic with estimated parameters. Biometrika 76(2), 385–392 (1989). doi:10.1093/biomet/76.2.385
Clarke, B. R. (2008). Linear Models : The Theory and Application of Analysis of Variance. http://UWA.eblib.com.au/patron/FullRecord.aspx?p=362059
Efron, B., Tibshirani, R.J.: An Introduction to the Boostrap. Chapman & Hall, New York (1993)
Freimer, M., Kollia, G., Mudholkar, G., Lin, C.: A Study of the generalised Tukey lambda family. Commun. Stat. 17(10), 3547–3567 (1988)
Gilchrist, W.: Modelling with quantile distributions. J. Appl. Stat. 24(1), 113–122 (1997)
Gilchrist, Warren: Statistical Modelling with Quantile Functions. Chapman and Hall/CRC, New York (2000)
Gilchrist, Warren: Regression revisited. Int. Stat. Rev. 76(3), 401–418 (2008)
Hyndman, R., Fan, Y.: Sample quantiles in statistical packages. Am. Stat. 50(4), 361–365 (1996)
Karian, Z., Dudewicz, E.: Fitting Statistical Distributions: The Generalized Lambda Distribution and Generalised Bootstrap Methods. Chapman and Hall, New York (2000)
Karvanen, J., Nuutinen, A.: Characterizing the generalized lambda distribution by L-moments. Comput. Stat. Data Anal. 52(4), 1971–1983 (2008)
King, R., MacGillivray, H.: A starship estimation method for the generalised lambda distributions. Aust. N. Z. J. Stat. 41(3), 353–374 (1999)
Koenker, R. (2013). Quantile Regression in R: A Vignette
Koenker, R.: Quantile Regression. Cambridge University Press, New York (2005)
Koenker, R., Bassett Jr, G.: Regression quantiles. Econometrica 46(1), 33–50 (1978)
Koenker, R., Bassett Jr, G.: Robust tests of heteroscedasticity based on regression quantiles. Econometrica 50, 43–61 (1982)
Lakhany, A, & Massuer, H. (2000). Estimating the parameters of the generalised lambda distribution. Algo Res. Q. 47–58.
Mosteller, F., Tukey, J.: Data Analysis and Regression: A Second Course in Statistics. Addison-Wesley, Reading (1977)
Parzen, E.: Nonparametric statistical data modeling. J. Am. Stat. Assoc. 74(365), 105–121 (1979)
Ramberg, J., Schmeiser, B.: An approximate method for generating asymmetric random variables. Commun. Assoc. Comput. Mach. 17, 78–82 (1974)
Ramberg, J., Tadikamalla, P., Dudewicz, E., Mykytka, E.: A probability distribution and its uses in fitting the data. Technometrics 21, 201–214 (1979)
Reich, B.J., Bondell, H.D., Wang, H.J.: Flexible Bayesian quantile regression for independent and clustered data. Biostatistics 11(2), 337–352 (2010). doi:10.1093/biostatistics/kxp049
Reich, B.J., Fuentes, M., Dunson, D.B.: Bayesian spatial quantile regression. J. Am. Stat. Assoc. 106(493), 6–20 (2011). doi:10.1198/jasa.2010.ap09237
Silverman, B.W.: Some aspects of the spline smoothing approach to non-parametric curve fitting. J. R. Stat. Soc. B 47, 1–52 (1985)
Su, S.: Distributional modelling of pipeline leakage repair costs for a water utility company. Am. J. Math. Manag. Sci. 27(3–4), 369–400 (2007a)
Su, S.: Fitting single and mixture of generalized lambda distributions to data via discretized and maximum likelihood methods: GLDEX. R. J. Stat. Softw. 21(9), 1–17 (2007b)
Su, S.: Numerical maximum log likelihood estimation for generalized lambda distributions. Comput. Stat. Data Anal. 51(8), 3983–3998 (2007c)
Su, S.: Chapter 14: Fitting GLD to data via quantile matching method. In: Karian, Z., Dudewicz, E. (eds.) Handbook of Distribution Fitting Methods with R, pp. 557–583. CRC Press/Taylor & Francis, Boca Raton (2010a)
Su, S.: Chapter 15: Fitting GLD to data using the GLDEX 1.0.4 in R. In: Karian, Z., Dudewicz, E. (eds.) Handbook of Distribution Fitting Methods with R (pp. 585–608). CRC Press/Taylor & Francis, Boca Raton (2010b)
Su, S.: Chapter 28: Distributional modeling of pipeline leakage repair costs for a water utility company. In: Karian, Z., Dudewicz, E. (eds.) Handbook of Distribution Fitting Methods with R, pp. 1171–1205. CRC Press/Taylor & Francis, Boca Raton (2010c)
Theil, H.: The analysis of disturbances in regression analysis. J. Am. Stat. Assoc. 60, 1067–1079 (1965)
Titterington, D.M.: Comment on ‘Estimating parameters in continuous univariate distributions. J. R. Stat. Soc. B 47(1), 115–116 (1985)
Yu, K., Moyeed, R.A.: Bayesian quantile regression. Stat. Probab. Lett. 54(4), 437–447 (2001)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Su, S. Flexible parametric quantile regression model. Stat Comput 25, 635–650 (2015). https://doi.org/10.1007/s11222-014-9457-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11222-014-9457-1