Efficient Estimation for Semi-varying Coefficient Model with An Invertible Linear Process Error

Abstract

For semi-varying-coefficient model with an invertible linear process error, we propose an efficient estimator procedure. This procedure is based on a pre-whitening transformation of the dependent variable that must be estimated from the data. We establish the proposed estimations’ asymptotic normalities, and assess their finite sample performance. Monte Carlo simulation suggest that the efficiency gain can be achieved in moderate-sized samples.

Keywords:

• Invertible linear process error
• Local linear regression
• Semivarying-coefficient model
• Whitening technique

Mathematics Subject Classification:

• Primary 62H12
• Secondary 62A10

Appendix

Table 1 The estimators of varying-coefficient function with n = 200

Table 2 The estimators of varying-coefficient function with n = 400

Table 3 The estimators of the parametric components with n = 200

Table 4 The estimators of the parametric components with n = 400