Review Articles

Estimation and inference for multi-kink expectile regression with nonignorable dropout

Dongyu Li ,

School of Statistics and Data Science, KLMDASR, LEBPS and LPMC, Nankai University, Tianjin, People's Republic of China

Lei Wang

School of Statistics and Data Science, KLMDASR, LEBPS and LPMC, Nankai University, Tianjin, People's Republic of China

Pages | Received 16 Jul. 2023, Accepted 03 Jan. 2024, Published online: 22 Jan. 2024,
  • Abstract
  • Full Article
  • References
  • Citations

In this paper, we consider parameter estimation, kink points testing and statistical inference for a longitudinal multi-kink expectile regression model with nonignorable dropout. In order to accommodate both within-subject correlations and nonignorable dropout, the bias-corrected generalized estimating equations are constructed by combining the inverse probability weighting and quadratic inference function approaches. The estimators for the kink locations and regression coefficients are obtained by using the generalized method of moments. A selection procedure based on a modified BIC is applied to estimate the number of kink points. We theoretically demonstrate the number selection consistency of kink points and the asymptotic normality of all estimators. A weighted cumulative sum type statistic is proposed to test the existence of kink effects at a given expectile, and its limiting distributions are derived under both the null and the local alternative hypotheses. Simulation studies show that the proposed estimators and test have desirable finite sample performance in both homoscedastic and heteroscedastic errors. An application to the Nation Growth, Lung and Health Study dataset is also presented.


  • Aigner, D., Amemiya, T., & Poirier, D. (1976). On the estimation of production frontiers: Maximum likelihood estimation of the parameters of a discontinuous density function. International Economic Review17(2), 377–396.
  • Bai, J., & Perron, P. (2003). Computation and analysis of multiple structural change models. Journal of Applied Econometrics18(1), 1–22.
  • Chen, J., Huang, Y., & Tang, N. (2022). Bayesian change-point joint models for multivariate longitudinal and time-to-event data. Statistics in Biopharmaceutical Research14(2), 227–241.
  • Fong, Y. (2019). Fast bootstrap confidence intervals for continuous threshold linear regression. Journal of Computational and Graphical Statistics28(2), 1–8.
  • Hansen, B. (2017). Regression kink with an unknown threshold. Journal of Business and Economic Statistics35(2), 228–240.
  • Hansen, L. (1982). Large sample properties of generalized method of moments estimators. Econometrica50(4), 1029–1054.
  • Koenker, R., & Bassett, J. (1978). Regression quantiles. Econometrica46(1), 33–50.
  • Lee, S., Seo, M., & Shin, Y. (2011). Testing for threshold effects in regression models. Journal of the American Statistical Association106(493), 220–231.
  • Lee, S., Seo, M., & Shin, Y. (2016). The lasso for high dimensional regression with a possible change point. Journal of the Royal Statistical Society: Series B (Statistical Methodology)78(1), 193–210.
  • Li, C., Wei, Y., Chappell, R., & He, X. (2011). Bent line quantile regression with application to an allometric study of land mammals' speed and mass. Biometrics67(1), 242–249.
  • Li, D., & Wang, L. (2022). Improved kth power expectile regression with nonignorable dropouts. Journal of Applied Statistics49(11), 2767–2788.
  • Li, D., Wang, L., & Zhao, W. (2022). Estimation and inference for multikink expectile regression with longitudinal data. Statistics in Medicine41(7), 1296–1313.
  • Liang, K., & Zeger, S. (1986). Longitudinal data analysis using generalized linear models. Biometrika73(1), 13–22.
  • Matteo, B., Haeran, C., & Piotr, F. (2018). Simultaneous multiple change-point and factor analysis for high dimensional time series. Journal of Econometrics206(1), 187–225.
  • Miao, W., & Tchetgen Tchetgn, E. (2018). Identification and inference with nonignorable missing covariate data. Statistica Sinica28(4), 2049–2067.
  • Newey, W., & Powell, J. (1987). Asymmetric least squares estimation and testing. Econometrica55(4), 819–847.
  • Oka, T., & Qu, Z. (2011). Estimating structural changes in regression quantiles. Journal of Econometrics162(2), 248–267.
  • Perron, P., & Qu, Z. (2006). Estimating restricted structural change models. Journal of Econometrics134(2), 373–399.
  • Qu, A., Lindsay, B., & Li, B. (2000). Improving generalised estimating equations using quadratic inference functions. Biometrika87(4), 823–836.
  • Shao, J., & Wang, L. (2016). Semiparametric inverse propensity weighting for nonignorable missing data. Biometrika103(1), 175–187.
  • Wan, C., Zhong, W., Zhang, W., & Zou, C. (2023). Multikink quantile regression for longitudinal data with application to progesterone data analysis. Biometrics79(2), 747–760.
  • Wang, L., Qi, C., & Shao, J. (2019). Model-assisted regression estimators for longitudinal data with nonignorable dropout. International Statistical Review87(S1), S121–S138.
  • Wang, S., Shao, J., & Kim, J. (2014). An instrumental variable approach for identification and estimation with nonignorable nonresponse. Statistica Sinica24(3), 1097–1116.
  • Zhang, F., & Li, Q. (2017). A continuous threshold expectile model. Computational Statistics and Data Analysis116, 49–66.
  • Zhang, L., Wang, J., & Zhu, Z. (2014). Testing for change points due to a covariate threshold in quantile regression. Statistica Sinica24(4), 1859–1877.
  • Zhong, W., Wan, C., & Zhang, W. (2022). Estimation and inference for multi-kink quantile regression. Journal of Business and Economic Statistics40(3), 1123–1139.
  • Zhou, H., & Liang, H. (2021). Change point estimation in regression model with response missing at random. Communications in Statistics-Theory and Methods51(20), 7101–7119.

To cite this article: Dongyu Li & Lei Wang (2024) Estimation and inference for multi-kink expectile regression with nonignorable dropout, Statistical Theory and Related Fields, 8:2, 136-151, DOI: 10.1080/24754269.2024.2302763

To link to this article: