Abstract
For the problem of estimating a regression function, $\\mu$ say,\nsubject to shape constraints, like monotonicity or convexity, it is argued that\nthe divergence of the maximum likelihood estimator provides a useful measure of\nthe effective dimension of the model. Inequalities are derived for the expected\nmean squared error of the maximum likelihood estimator and the expected\nresidual sum of squares. These generalize equalities from the case of linear\nregression. As an application, it is shown that the maximum likelihood\nestimator of the error variance $\\sigma^2$ is asymptotically normal with mean\n$\\sigma^2$ and variance $2\\sigma_2/n$. For monotone regression, it is shown\nthat the maximum likelihood estimator of $\\mu$ attains the optimal rate of\nconvergence, and a bias correction to the maximum likelihood estimator of\n$\\sigma^2$ is derived.
Keywords
MathematicsEstimatorStatisticsMean squared errorBias of an estimatorMinimum-variance unbiased estimatorApplied mathematicsConvexityRestricted maximum likelihoodConsistent estimatorRate of convergenceMonotonic functionM-estimatorEstimation theoryMathematical analysis
Related Publications
A Nondegenerate Penalized Likelihood Estimator for Variance Parameters in Multilevel Models
Group-level variance estimates of zero often arise when fitting multilevel or hierarchical linear models, especially when the number of groups is small. For situations where zer...
Estimation in a Multivariate "Errors in Variables" Regression Model: Large Sample Results
In a multivariate "errors in variables" regression model, the unknown mean vectors $\\mathbf{u}_{1i}: p \\times 1, \\mathbf{u}_{2i}: r \\times 1$ of the vector observations $\\m...
Theory for penalised spline regression
Penalised spline regression is a popular new approach to smoothing, but its theoretical properties are not yet well understood. In this paper, mean squared error expressions and...
Analyzing bagging
Bagging is one of the most effective computationally intensive procedures to improve on unstable estimators or classifiers, useful especially for high dimensional data set probl...
Asymptotically Efficient Selection of the Order of the Model for Estimating Parameters of a Linear Process
Let $\\{x_t\\}$ be a linear stationary process of the form $x_t + \\Sigma_{1\\leqslant i<\\infty}a_ix_{t-i} = e_t$, where $\\{e_t\\}$ is a sequence of i.i.d. normal random va...
Publication Info
- Year
- 2000
- Type
- article
- Volume
- 28
- Issue
- 4
- Citations
- 154
- Access
- Closed
External Links
Social Impact
Altmetric
PlumX Metrics
Social media, news, blog, policy document mentions
Citation Metrics
154
OpenAlex
Cite This
Mary C. Meyer,
Michael Woodroofe
(2000).
On the degrees of freedom in shape-restricted regression.
The Annals of Statistics
, 28
(4)
.
https://doi.org/10.1214/aos/1015956708
Identifiers
- DOI
- 10.1214/aos/1015956708