Parametrically Guided Non‐parametric Regression
The problem of the conditional mean function \(m(x)=E(Y | X=x)\) nonparametric estimation by the observation of i.i.d. pairs \((X_i,Y_i)\) is considered. It is proposed to use a parametrically guided estimator, based on the identity \(m(x)=m(x,\beta)r(x)\), where \(m(x,\beta)\) is some ``parametric guide for \(m\) which is fitted by an appropriate choice of the parameter \(\beta\) and \(r(x)=m(x)/m(x,\beta)\) is the correction factor. It is proposed to use the maximum likelihood estimator \(\hat\beta\) for \(\beta\) (in the Gaussian case this estimator estimates \(\beta_0\) which minimizes the \(L_2\) distance between \(m(x)\) and \(m(x,\beta)\)). The correction factor is estimated by a kernel weighted local polynomial regression estimator \(\hat r(x)\) of degree \(p\) based on the ``observations \((X_i, Y_i | m(X_i,\hat\beta))\). The leading term of bias and variance of the estimator \(\hat m(x)=m(x,\hat\beta)\hat r(x)\), under some regularity conditions, is obtained. The leading term of the variance is the same as for the local polynomial estimator \(m^*\) with \(m(x,\beta)=1\) and the leading term of the bias for \(\hat m\) is better for \(m^*\) if \(| m(x,\beta_0)r(x)| <| m(x)| \) (in the case \(p=1\)). Some examples are considered and simulation results are presented.
- A note on unconditional properties of a parametrically guided Nadaraya-Watson estimator
- \(\sqrt{n}\)-consistent density estimation in semiparametric regression models
- Parametrically guided nonparametric density and hazard estimation with censored data
- Asymptotic normality of a combined regression estimator
- Bias reduction by projection on parametric models in Hilbertian nonparametric regression
- Nonparametric regression with parametric help
- Finite population model-assisted estimation using combined parametric and nonparametric regression smoothers
- Parametrically guided local quasi-likelihood with censored data
- Nonparametric and semiparametric regressions subject to monotonicity constraints: estimation and forecasting
- Nonparametric regression under alternative data environments
- A Modified Nonparametric Prewhitened Covariance Estimator
- A Class of Improved Parametrically Guided Nonparametric Regression Estimators
- An Adaptive Two‐stage Estimation Method for Additive Models
- Multiplicative adjustment method for semiparametric regression with mixing dependent data
- On hybrid classification using model assisted posterior estimates
- SEMIPARAMETRIC REGRESSION WITH MULTIPLICATIVE ADJUSTMENT
- A semiparametric conditional duration model
- Semiparametric multiple kernel estimators and model diagnostics for count regression functions
- Guided Censored Regression
- Parametrically guided estimation in nonparametric varying coefficient models with quasi-likelihood
- Parametrically guided generalised additive models with application to mergers and acquisitions data
- Hilbertian additive regression with parametric help
- Can a Machine Correct Option Pricing Models?
- Consistent inference for biased sub-model of high-dimensional partially linear model
- Structure discovery and parametrically guided regression
- An iterative approach for deriving and solving an accurate regression equation
- Nonparametric estimation of densities on the hypersphere using a parametric guide
- Additive regression with parametric help
- Varying coefficient regression: revisit and parametric help
- Generalized parametric help in Hilbertian additive regression
- Simulation-based consistent inference for biased working model of non-sparse high-dimensional linear regression
- Semiparametric estimation of the link function in binary-choice single-index models
- Bayesian estimation of bandwidth in semiparametric kernel estimation of unknown probability mass and regression functions of count data
- Nonparametric prediction of stock returns based on yearly data: the long-term view
This page was built for publication: Parametrically Guided Non‐parametric Regression
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4702167)
