Abstract
The collection rate of contributions to public pension (CRCP), expressed as the ratio of the actual contributions to the expected contributions from insurers, is a key component of the public pension system in China. Recent years have seen various patterns of change in CRCPs at the provincial level. In order to study the drastic changes in a short time and understand their underlying implications, we propose a nonparametric time-varying coefficients model for longitudinal data with pre-specified finite time points, also known as panel data. By utilizing a penalized least squares method, the proposed method enables estimation of a large number of parameters, which can exceed the sample size. The resulting estimator is shown to be efficient, robust, and computationally feasible. Furthermore, it possesses desirable theoretical properties such as \(n^{1/2}\)-consistency, asymptotic normality, and the oracle property.
Similar content being viewed by others
References
Cai Z (2007) Trending time-varying coefficient time series models with serially correlated errors. J Economet 136:163–188
Cai Z, Sun Y (2003) Local linear estimation for time-dependent coefficients in Cox’s regression models. Scand J Stat 30:93–111
Cai Z, Fan J, Yao Q (2000) Functional-coefficient regression models for nonlinear time series models. J Am Stat Assoc 95:941–956
Chen R, Tsay RS (1993) Functional-coefficient autoregressive models. J Am Stat Assoc 88:298–308
Chen K, Lin H, Zhou Y (2012) Efficient estimation for the Cox model with varying coefficients. Biometrika 99:379–392
Efron B, Hastie T, Johnstone I, Tibshirani R (2004) Least angle regression (with discussion). Ann Stat 32:407–499
Fan J, Gijbels I (1996) Local polynomial modeling and its applications. Chapman and Hall, London
Fan J, Zhang W (1999) Statistical estimation in varying coefficient models. Ann Stat 27:1491–1518
Fan J, Li R (2001) Variable selection via nonconcave penalized likelihood and its oracle properties. J Am Stat Assoc 96:1348–1360
Fan J, Yao Q (2003) Nonlinear time series: nonparametric and parametric methods. Springer, New York
Fan J, Zhang JT (2000) Two-Step estimation of functional linear models with applications to longitudinal data. J R Stat Soc B 62:303–322
Fan J, Lin H, Zhou Y (2006) Local partial likelihood estimation for life time data. Ann Stat 34:290–325
Fan J, Huang T, Li R (2007) Analysis of longitudinal data with semiparametric estimation of covariance function. J Am Stat Assoc 102:632–641
Feng J, He L, Satob H (2011) Public pension and household saving: evidence from urban China. J Comp Econ 39:470–485
Friedman J, Hastie T, Tibshirani R (2001) The elements of statistical learning. Springer series in statistics. Springer, New York
Gamerman D (1991) Markov chain Monte Carlo for dynamic generalized linear models. Biometrika 85:215–227
Gao Q (2010) Redistributive nature of the Chinese social benefit system: progressive or regressive? China Q 201:1–19
Gillion C (2000) Social security pensions: development and reform. International Labour Organisation, Geneva
Hastie T, Tibshirani R (1990) Generalized additive models. Chapman and Hall, London
Hastie T, Tibshirani R (1993) Varying-coefficient models (with discussion). J R Stat Soc B 55:757–796
Hess W, Persson M, Rubenbauer S, Gertheiss J (2013) Using lasso-type penalties to model time-varying covariate effects in panel data regressions-a novel approach illustrated by “Death of Distance” in international trade. Working Paper
Hoover DR, Rice JA, Wu CO, Yang LP (1998) Nonparametric smoothing estimates of time-varying coefficient models with longitudinal data. Biometrika 85:809–822
Huang J, Shen H (2004) Functional coefficient regression models for non-linear time series: a polynomial spline approach. Scand J Stat 31:515–534
Hunter DR, Li R (2005) Variable selection using MM algorithms. Ann Stat 33:1617–1642
Li DG, Chen J, Gao JT (2011) Non-parametric time-varying coefficient panel data models with fixed effects. Econ J 14:387–408
Lin DY, Ying Z (2001) Semiparametric and nonparametric regression analysis of longitudinal data (with discussion). J Am Stat Assoc 96:103–113
Lin H, Peng H (2013) Smoothed rank correlation of the linear transformation regression model. Comput Stat Data Anal 57(1):615–630
Lin H, Song XK, Zhou Q (2007) Varying-coefficient marginal models and applications in longitudinal data analysis. Sankhya 69:581–614
Lin H, Zhou L, Peng H, Zhou XH (2011) Selection and combination of biomarkers using ROC method for disease classification and prediction. Can J Stat 39(2):324–343
Liu J (2011) Resources, incentives and sectoral interests: a longitudinal study of the system of collecting social insurance contributions in China (1999–2008). Soc Sci China 3:9
Martinussen T, Scheike TH, Skovgaard IM (2000) Efficient estimation of fixed and time-varying covariates effects in multiplicative intensity models. Scand J Stat 29:57–74
Marzec L, Marzec P (1997) On fitting Cox’s regression model with time-dependent coefficients. Biometrika 84:901–908
Murphy SA (1993) Testing for a time dependent coefficient in Cox’s regression model. Scand J Stat 20:35–50
Murphy SA, Sen PK (1991) Time-dependent coefficients in a Cox-type regression model. Stoch Process Appl 39(1):153–180
Nielsen I, Smyth R (2008a) Job satisfaction and response to incentives among China’s urban workforce. J Socio Econ 37:1921–1936
Nielsen I, Smyth R (2008b) Who bears the burden of employer compliance with social security contributions? Evidence from Chinese firm level data. China Econ Rev 19:230–244
Nyland C, Smyth R, Zhu J (2006) What determines the extent to which employers will comply with their social security obligations? Evidence from Chinese firm-level data. Soc Policy Admin 40:196–214
Olsen MK, Schafer J (2001) A two-part random-effects model for semi-continuous longitudinal data. J Am Stat Assoc 96:730–745
Orbe S, Ferreira E, Rodriguez-Poo J (2005) Nonparametric estimation of time varying parameters under shape restrictions. J Economet 126:53–57
Palacios R, Pallares-Miralles M (2000) International patterns of pension provision. Social Protection Discussion Paper Series No. 0009. The World Bank, Washington, DC
Phillips P (2001) Trending time series and macroeconomic activity: some present and future challenges. J Economet 100:21–27
Qian J, Wang L (2012) Estimating semiparametric panel data models by marginal integration. J Economet 167:483–493
Queisser M, Reilly A, Hu Y (2016) China’s pension system and reform: an OECD perspective. Econ Political Stud 4:345–367
Ramsay JO, Silverman BW (1997) Functional data analysis. Springer, New York
Roberts S, Stafford B, Ashworth, K (2004) Assessing the coverage gap. ISSA initiative findings and opinions. 12
Robinson PM (1989) Nonparametric estimation of time-varying parameters. Statistical analysis and forecasting of economic structural change. Springer, Berlin, pp 253–264
Robinson PM (1991) Time-varying nonlinear regression. Economic structure change. Springer, Berlin, pp 179–190
Robinson PM (2012) Nonparametric trending regression with cross-sectional dependence. J Economet 169(1):4–14
Rodriguez-Poo J, Soberon A (2014) Direct semi-parametric estimation of fixed effects panel data varying coefficient models. Econ J 17:107–138
Stanovnik T, Bejakovic P, Chlon-Dominczak A (2015) The collection of pension contributions: a comparative review of three Central European countries. Econ Res Ekon Istraz 28:1149–1161
Sun YG, Carroll RJ, Li DD (2009) Semiparametric estimation of fixed effects panel data varying coeffcient models. Adv Econom 25:101–129
Tian L, Zucker D, Wei LJ (2005) On the Cox model with time-varying regression coefficients. J Am Stat Assoc 100:172–183
Tibshirani R (1996) Regression shrinkage and selection via the LASSO. J R Stat Soc B 58:267–288
Wang H, Li R, Tsai C (2007) Tuning parameter selectors for the smoothly clipped absolute deviation method. Biometrika 94:553–568
Wu CO, Chiang CT, Hoover DR (1998) Asymptotic confidence regions for kernel smoothing of a varying coefficient model with longitudinal data. J Am Stat Assoc 93:1388–1402
Zhang CH (2010) Nearly unbiased variable selection under minimax concave penalty. Ann Stat 38:894–942
Zou H (2006) The adaptive Lasso and its oracle properties. J Am Stat Assoc 101:1418–1429
Zou H, Hastie T (2005) Regularization and variable selection via the elastic net. J R Stat Soc B 67(2):301–320
Zou H, Li R (2008) One-step sparse estimates in nonconcave penalized likelihood models. Ann Stat 36:1509–1533
Zucker DM, Karr AF (1990) Nonparametric survival analysis with time-dependent covariate effects: a penalized partial likelihood approach. Ann Stat 18:329–353
Acknowledgements
We thank the Editor, the AE, and two referees for the helpful suggestions that helped improve much the manuscript. The research is partially supported by the National Natural Science of China (No. 11829101, 11571282) and the Fundamental Research Funds for the Central Universities (JBK120509, JBK140507).
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix
Proof of Theorem 1
Let \(\alpha _n=n^{-1/2}+a_n\). Denote by \(\theta _0\) the true value of \(\theta \). We want to show that for any given \(\varepsilon >0\), there exists a large constant C such that
This implies with a probability larger than \(1-\varepsilon \) that there exists a local minimum in the ball \( \{\theta _0+ \alpha _n \cdot u: \Vert u\Vert \le C\}.\) Hence, there exists a local minimizer such that \(\Vert {\widehat{\theta }}-\theta _0\Vert =O_p(\alpha _n)\).
Define \(\theta ^{*}=\theta _0 + \alpha _n \cdot u=(\theta _1^{*},\ldots ,\theta _{Tp+T-1}^{*})^{\prime }\), using \(p_{\lambda }(0)=0\), we have
where \(S(\theta )=\frac{1}{n}\sum _{i=1}^{n}\sum _{t>s}\Big \{Y_{it} -Y_{is}-\sum _{d=s+1}^t g_d-\sum _{j=1}^p\Big (\sum _{d=1}^t \gamma _{dj}X_{it,j}-\sum _{d=1}^s \gamma _{dj}X_{is,j}\Big ) \Big \}^2\), m is the number of components of \(\theta ^{(1)}_0\). Let \({\dot{S}}\) be the gradient vector of S; by the standard argument of the Taylor expansion, we have
Noting that \(E\{{\dot{S}}(\theta _0)\}=0\) and \(Var\{{\dot{S}}(\theta _0)\}=O(n^{-1})\), by the central limit theory we have
Similarly, we get
For \(I_3\), it is easy to see that it is bounded by
From (5.3), (5.4), and (5.5), \(I_1\) and \(I_3\) are dominated by \(I_2\). Hence, by choosing a sufficiently large C, (5.1) holds. \(\square \)
Proof of Theorem 2
We first show that with a probability tending to 1, for any given \(\theta ^{(1)}\) satisfying \(\Vert \theta ^{(1)}-\theta ^{(1)}_0\Vert =O_p(n^{-1/2})\) and any constant C,
To show (5.6), by Taylor’s expansion, we have
By the central limit theorem, we have
where \(\liminf \limits _{n\rightarrow \infty }\liminf \limits _{\theta \rightarrow 0^{+}} \lambda ^{-1}{\dot{p}}_{\lambda }(\theta )>0\) and \(n^{-1/2}/\lambda \rightarrow 0\). The sign of the derivative is completely determined by that of \(\theta _r\). Hence (5.6) follows.
By (5.6), Part (a) follows. Now we prove Part (b). It can be shown that there exists \({\widehat{\theta }}^{(1)}\) in Theorem 1 that is a \(n^{1/2}\)- consistent local maximizer of \(L_n(({\theta ^{(1)}}', 0')')\), which is regarded as a function of \(\theta ^{(1)}\), and that satisfies the following equation:
Note that \({\widehat{\theta }}^{(1)}\) is a constant estimator. Thus, we have
Furthermore, we have
where
Hence following by Slutsky’s theorem we have
This completes the proof of Part (b). \(\square \)
Rights and permissions
About this article
Cite this article
Lin, H., Hong, H.G., Yang, B. et al. Nonparametric Time-Varying Coefficient Models for Panel Data. Stat Biosci 11, 548–566 (2019). https://doi.org/10.1007/s12561-019-09248-0
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12561-019-09248-0