On bias reduction methods in nonparametric regression estimation
Let i.i.d. \((Y_i,x_i)\), \(i=1,\dots,n\), be observed where \(x_i\) are independent, and \(Y_i\) are dependent variables. The nonparametric local linear fitting estimator (NLLE) for the regression curve is defined as \(\hat m_h^{LL}(x)=a_0+a_1 x,\) where \(a_0=a_0(x)\) and \(a_1=a_1(x)\) are chosen to minimize \({1\over{h}}\sum_{i=1}^n K\left({{x-x_i}\over{h}}\right)(Y_i-a_0-a_1x)^2,\) and \(K\) is some weight function (kernel). Two methods of bias reduction are considered for \(\hat m_n^{LL}\). In the multiplicative correction method the corrected estimator is \(\hat m^{JLN}_h=\hat m^{LL}_h(x)\hat r_0(x)\), where \(\hat r_0(x)\) is the NLLE for the data \((x_i, Y_i/\hat m_h^{LL}(x_i))\). In the transformation method the data \((Y_i,\hat m_h^{LL}(x_i))\) are fitted by the NLLE and the result is transformed to the original scale. It is known that these methods reduce bias from \(O(h^2)\) to \(O(h^4)\). Results of Monte-Carlo studies are presented which show that these methods outperform NLLE for \(n=400\). For the entire collection see [Zbl 0990.62506].
- A bias corrected nonparametric regression estimator
- Bias reduction for nonparametric and semiparametric regression models
- A multiplicative bias reduction method for nonparametric regression
- On bias reduction in local linear smoothing
- Simple Transformation Techniques for Improved Non‐parametric Regression
- Double-smoothing for bias reduction in local linear regression
- A multiplicative bias reduction method for nonparametric regression
- The efficiency of bias-corrected estimators for nonparametric kernel estimation based on local estimating equations
- Bias reduction by projection on parametric models in Hilbertian nonparametric regression
- A bias corrected nonparametric regression estimator
- On bias reduction in local linear smoothing
- scientific article; zbMATH DE number 635765 (Why is no real title available?)
- Simple Transformation Techniques for Improved Non‐parametric Regression
- Bias reduction for nonparametric and semiparametric regression models
- Bias and variance reduction procedures in non-parametric regression
- Identity reproducing multivariate nonparametric regression
This page was built for publication: On bias reduction methods in nonparametric regression estimation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3843169)
