M‐estimation Under a Two‐Sample Semiparametric Model
The author considers data which consist of two i.i.d. samples \(X_1\),\dots, \(X_{n_0}\sim g(x)\), \(Z_1\),\dots, \(Z_{n_1}\sim h(x)=\exp[\alpha+r(x)\beta]g(x)\), where \(g(x)\) is the PDF of some unknown distribution, \(r(x)\) is a \(p\)-dimensional vector of known functions, and \(\alpha\) and \(\beta=(\beta_1,\dots,\beta_p)^T\) are unknown parameters. The problem is to estimate some unknown parameter \(\vartheta=\vartheta(G)\). It is proposed to use a two-stage estimation procedure. At the first stage \(G\), \(\alpha\) and \(\beta\) are estimated by an empirical maximum likelihood estimator \(\tilde G_n\), \(\tilde\alpha_n\), \(\tilde\beta_n\) (\(n=n_0+n_1\)). At the second stage \(\vartheta\) is estimated by some M-estimator as a solution of the estimating equation \( \int\psi(x,\vartheta)d\tilde G_n(x)=0 \), where \(\psi\) is a function such that \( \int\psi(x,\vartheta_0)d G(x)=0 \) for the true value of the parameter \(\vartheta_0=\vartheta(G)\). Consistency and asymptotic normality of these estimators are demonstrated. It is shown that they are more efficient (asymptotically) than M-estimators constructed by \(X_i\) only. A plug-in estimator for the limit variance is constructed and its consistency is demonstrated. Simulation results are presented.
- Efficient estimation in the two-sample semiparametric location-scale models
- scientific article; zbMATH DE number 4201378
- Minimum Hellinger distance estimation in a two-sample semiparametric model
- Rank estimates in a class of semiparametric two-sample models
- Quantile estimation under a two-sample semi-parametric model
- A simplified approach to M-estimation with application to two-stage estimators
- Semiparametric density estimation under a two-sample density ratio model
- Efficient estimation in the two-sample semiparametric location-scale models
- An overview of nonparametric contributions to the problem of functional estimation from biased data
- Comparing two samples by penalized logistic regression
- Estimation in two sample randomly truncated scale model
- Prospective and retrospective analyses under logistic regression models
- Forecasting mortality rates via density ratio modeling
- Plug-In Two-Stage Normal Density Estimation Under MISE Loss: Unknown Variance
- scientific article; zbMATH DE number 1522488 (Why is no real title available?)
- Asymptotic Confidence Intervals Based on M-procedures in One- and Two-sample Models
- scientific article; zbMATH DE number 795285 (Why is no real title available?)
- Two-step combined nonparametric likelihood estimation of misspecified semiparametric models
- Using entropy distance measures for testing odds ratio parameters under logistic regression models based on case-control data
- Rank estimates in a class of semiparametric two-sample models
- Semiparametric distribution forecasting
This page was built for publication: M‐estimation Under a Two‐Sample Semiparametric Model
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4516151)
