Non-parametric inference for density modes
From MaRDI portal
Abstract: We derive nonparametric confidence intervals for the eigenvalues of the Hessian at modes of a density estimate. This provides information about the strength and shape of modes and can also be used as a significance test. We use a data-splitting approach in which potential modes are identified using the first half of the data and inference is done with the second half of the data. To get valid confidence sets for the eigenvalues, we use a bootstrap based on an elementary-symmetric-polynomial (ESP) transformation. This leads to valid bootstrap confidence sets regardless of any multiplicities in the eigenvalues. We also suggest a new method for bandwidth selection, namely, choosing the bandwidth to maximize the number of significant modes. We show by example that this method works well. Even when the true distribution is singular, and hence does not have a density, (in which case cross validation chooses a zero bandwidth), our method chooses a reasonable bandwidth.
Recommendations
Cited in
(23)- Nonparametric testing of the existence of modes
- Posterior contraction rates of density derivative estimation
- Counting bumps
- Nonparametric semi-supervised classification with application to signal detection in high energy physics
- Asymptotics for function derivatives estimators based on stationary and ergodic discrete time processes
- Estimation of the global mode of a density: minimaxity, adaptation, and computational complexity
- Optimal adaptive estimation on \(\mathbb{R}\) or \(\mathbb{R}^{+}\) of the derivatives of a density
- Uniform convergence rate of the kernel regression estimator adaptive to intrinsic dimension in presence of censored data
- Uniform almost sure convergence and asymptotic distribution of the wavelet-based estimators of partial derivatives of multivariate density function under weak dependence
- Some results about kernel estimators for function derivatives based on stationary and ergodic continuous time processes with applications
- Bandwidth selection for nonparametric modal regression
- Learning by unsupervised nonlinear diffusion
- The Modal Age of Statistics
- Nonparametric needlet estimation for partial derivatives of a probability density function on the d -torus
- Persistence Flamelets: Topological Invariants for Scale Spaces
- Modal clustering of matrix-variate data
- Nonparametric recursive estimation for multivariate derivative functions by stochastic approximation method
- Moving Up the Cluster Tree with the Gradient Flow
- Bayesian taut splines for estimating the number of modes
- Locating \(\gamma\)-ray sources on the celestial sphere via modal clustering
- A fresh look at mean-shift based modal clustering
- Clustering by Hill-climbing: consistency results
- Random irregular histograms
This page was built for publication: Non-parametric inference for density modes
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5743227)
