CLT of wavelet estimator in semiparametric model with correlated errors

This paper considers the semiparametric regression model y _i = x _i ??+g(t _i )+V _i (1 ?? i ?? n), where (x _i , t _i ) are known design points, ?? is an unknown slope parameter, g(·) is an unknown function, the correlated errors $V_i = sumnolimits_{j = - infty }^infty {c_j e_{i - j} } $ with $sumnolimits_{j = - infty }^infty {|c_j |} < infty $ , and e _i are negatively associated random variables. Under appropriate conditions, the authors study the asymptotic normality for wavelet estimators of ?? and g(·). A simulation study is undertaken to investigate finite sample behavior of the estimators.