arXiv subjects

Larry Wasserman

Publications and source records attributed to Larry Wasserman.

At least 91 records · Page 5Linked to original sources

arXiv2018

Hybrid Wasserstein Distance and Fast Distribution Clustering

We define a modified Wasserstein distance for distribution clustering which inherits many of the properties of the Wasserstein distance but which can be estimated easily and computed quickly. The modified distance is the sum of two terms. The first term --- which has a closed form --- measures the location-scale differences between the distributions. The second term is an approximation that measures the remaining distance after accounting for location-scale differences. We consider several forms of approximation with our main emphasis being a tangent space approximation that can be estimated using nonparametric regression. We evaluate the strengths and weaknesses of this approach on simulated and real examples.

stat.ME↗

arXiv2018

Least Ambiguous Set-Valued Classifiers with Bounded Error Levels

In most classification tasks there are observations that are ambiguous and therefore difficult to correctly label. Set-valued classifiers output sets of plausible labels rather than a single label, thereby giving a more appropriate and informative treatment to the labeling of ambiguous instances. We introduce a framework for multiclass set-valued classification, where the classifiers guarantee user-defined levels of coverage or confidence (the probability that the true label is contained in the set) while minimizing the ambiguity (the expected size of the output). We first derive oracle classifiers assuming the true distribution to be known. We show that the oracle classifiers are obtained from level sets of the functions that define the conditional probability of each class. Then we develop estimators with good asymptotic and finite sample properties. The proposed estimators build on existing single-label classifiers. The optimal classifier can sometimes output the empty set, but we provide two solutions to fix this issue that are suitable for various practical needs.

stat.ME↗

arXiv2018

Analysis of a Mode Clustering Diagram

Mode-based clustering methods define clusters to be the basins of attraction of the modes of a density estimate. The most common version is mean shift clus- tering which uses a gradient ascent algorithm to find the basins. Rodriguez and Laio (2014) introduced a new method that is faster and simpler than mean shift clustering. Furthermore, they define a clustering diagram that provides a sim- ple, two-dimensional summary of the mode clustering information. We study the statistical properties of this diagram and we propose some improvements and extensions. In particular, we show a connection between the diagram and robust linear regression.

stat.ME↗

arXiv2018

Bootstrapping and Sample Splitting For High-Dimensional, Assumption-Free Inference

Several new methods have been proposed for performing valid inference after model selection. An older method is sampling splitting: use part of the data for model selection and part for inference. In this paper we revisit sample splitting combined with the bootstrap (or the Normal approximation). We show that this leads to a simple, assumption-free approach to inference and we establish results on the accuracy of the method. In fact, we find new bounds on the accuracy of the bootstrap and the Normal approximation for general nonlinear parameters with increasing dimension which we then use to assess the accuracy of regression inference. We show that an alternative, called the image bootstrap, has higher coverage accuracy at the cost of more computation. We define new parameters that measure variable importance and that can be inferred with greater accuracy than the usual regression coefficients. There is a inference-prediction tradeoff: splitting increases the accuracy and robustness of inference but can decrease the accuracy of the predictions.

math.ST↗

arXiv2017

Hypothesis Testing for High-Dimensional Multinomials: A Selective Review

The statistical analysis of discrete data has been the subject of extensive statistical research dating back to the work of Pearson. In this survey we review some recently developed methods for testing hypotheses about high-dimensional multinomials. Traditional tests like the $χ^2$ test and the likelihood ratio test can have poor power in the high-dimensional setting. Much of the research in this area has focused on finding tests with asymptotically Normal limits and developing (stringent) conditions under which tests have Normal limits. We argue that this perspective suffers from a significant deficiency: it can exclude many high-dimensional cases when - despite having non Normal null distributions - carefully designed tests can have high power. Finally, we illustrate that taking a minimax perspective and considering refinements of this perspective can lead naturally to powerful and practical tests.

stat.ML↗

arXiv2017

Uniform Asymptotic Inference and the Bootstrap After Model Selection

Recently, Tibshirani et al. (2016) proposed a method for making inferences about parameters defined by model selection, in a typical regression setting with normally distributed errors. Here, we study the large sample properties of this method, without assuming normality. We prove that the test statistic of Tibshirani et al. (2016) is asymptotically valid, as the number of samples n grows and the dimension d of the regression problem stays fixed. Our asymptotic result holds uniformly over a wide class of nonnormal error distributions. We also propose an efficient bootstrap version of this test that is provably (asymptotically) conservative, and in practice, often delivers shorter intervals than those from the original normality-based approach. Finally, we prove that the test statistic of Tibshirani et al. (2016) does not enjoy uniform validity in a high-dimensional setting, when the dimension d is allowed grow.

math.ST↗

arXiv2017

Hypothesis Testing For Densities and High-Dimensional Multinomials: Sharp Local Minimax Rates

We consider the goodness-of-fit testing problem of distinguishing whether the data are drawn from a specified distribution, versus a composite alternative separated from the null in the total variation metric. In the discrete case, we consider goodness-of-fit testing when the null distribution has a possibly growing or unbounded number of categories. In the continuous case, we consider testing a Lipschitz density, with possibly unbounded support, in the low-smoothness regime where the Lipschitz parameter is not assumed to be constant. In contrast to existing results, we show that the minimax rate and critical testing radius in these settings depend strongly, and in a precise way, on the null distribution being tested and this motivates the study of the (local) minimax rate as a function of the null distribution. For multinomials the local minimax rate was recently studied in the work of Valiant and Valiant. We re-visit and extend their results and develop two modifications to the chi-squared test whose performance we characterize. For testing Lipschitz densities, we show that the usual binning tests are inadequate in the low-smoothness regime and we design a spatially adaptive partitioning scheme that forms the basis for our locally minimax optimal tests. Furthermore, we provide the first local minimax lower bounds for this problem which yield a sharp characterization of the dependence of the critical radius on the null hypothesis being tested. In the low-smoothness regime we also provide adaptive tests, that adapt to the unknown smoothness parameter. We illustrate our results with a variety of simulations that demonstrate the practical utility of our proposed tests.

math.ST↗

arXiv2017

Statistical Inference using the Morse-Smale Complex

The Morse-Smale complex of a function $f$ decomposes the sample space into cells where $f$ is increasing or decreasing. When applied to nonparametric density estimation and regression, it provides a way to represent, visualize, and compare multivariate functions. In this paper, we present some statistical results on estimating Morse-Smale complexes. This allows us to derive new results for two existing methods: mode clustering and Morse-Smale regression. We also develop two new methods based on the Morse-Smale complex: a visualization technique for multivariate functions and a two-sample, multivariate hypothesis test.

math.ST↗

arXiv2017

Distribution-Free Predictive Inference For Regression

We develop a general framework for distribution-free predictive inference in regression, using conformal inference. The proposed methodology allows for the construction of a prediction band for the response variable using any estimator of the regression function. The resulting prediction band preserves the consistency properties of the original estimator under standard assumptions, while guaranteeing finite-sample marginal coverage even when these assumptions do not hold. We analyze and compare, both empirically and theoretically, the two major variants of our conformal framework: full conformal inference and split conformal inference, along with a related jackknife method. These methods offer different tradeoffs between statistical accuracy (length of resulting prediction intervals) and computational efficiency. As extensions, we develop a method for constructing valid in-sample prediction intervals called {\it rank-one-out} conformal inference, which has essentially the same computational efficiency as split conformal inference. We also describe an extension of our procedures for producing prediction bands with locally varying length, in order to adapt to heteroskedascity in the data. Finally, we propose a model-free notion of variable importance, called {\it leave-one-covariate-out} or LOCO inference. Accompanying this paper is an R package {\tt conformalInference} that implements all of the proposals we have introduced. In the spirit of reproducibility, all of our empirical results can also be easily (re)generated using this package.

stat.ME↗

arXiv2017

Statistical Inference for Cluster Trees

A cluster tree provides a highly-interpretable summary of a density function by representing the hierarchy of its high-density clusters. It is estimated using the empirical tree, which is the cluster tree constructed from a density estimator. This paper addresses the basic question of quantifying our uncertainty by assessing the statistical significance of topological features of an empirical cluster tree. We first study a variety of metrics that can be used to compare different trees, analyze their properties and assess their suitability for inference. We then propose methods to construct and summarize confidence sets for the unknown true cluster tree. We introduce a partial ordering on cluster trees which we use to prune some of the statistically insignificant features of the empirical tree, yielding interpretable and parsimonious cluster trees. Finally, we illustrate the proposed methods on a variety of synthetic examples and furthermore demonstrate their utility in the analysis of a Graft-versus-Host Disease (GvHD) data set.

math.ST↗

arXiv2017

Detecting Effects of Filaments on Galaxy Properties in the Sloan Digital Sky Survey III

We study the effects of filaments on galaxy properties in the Sloan Digital Sky Survey (SDSS) Data Release 12 using filaments from the `Cosmic Web Reconstruction' catalogue (Chen et al. 2016), a publicly available filament catalogue for SDSS. Since filaments are tracers of medium-to-high density regions, we expect that galaxy properties associated with the environment are dependent on the distance to the nearest filament. Our analysis demonstrates that a red galaxy or a high-mass galaxy tend to reside closer to filaments than a blue or low-mass galaxy. After adjusting the effect from stellar mass, on average, early-forming galaxies or large galaxies have a shorter distance to filaments than late-forming galaxies or small galaxies. For the Main galaxy sample (MGS), all signals are very significant ($>6σ$). For the LOWZ and CMASS sample, the stellar mass and size are significant ($>2 σ$). The filament effects we observe persist until $z = 0.7$ (the edge of the CMASS sample). Comparing our results to those using the galaxy distances from redMaPPer galaxy clusters as a reference, we find a similar result between filaments and clusters. Moreover, we find that the effect of clusters on the stellar mass of nearby galaxies depends on the galaxy's filamentary environment. Our findings illustrate the strong correlation of galaxy properties with proximity to density ridges, strongly supporting the claim that density ridges are good tracers of filaments.

astro-ph.GA↗

arXiv2016

Topological Data Analysis

Topological Data Analysis (TDA) can broadly be described as a collection of data analysis methods that find structure in data. This includes: clustering, manifold estimation, nonlinear dimension reduction, mode estimation, ridge estimation and persistent homology. This paper reviews some of these methods.

stat.ME↗

arXiv2016

Density Level Sets: Asymptotics, Inference, and Visualization

We derive asymptotic theory for the plug-in estimate for density level sets under Hausdoff loss. Based on the asymptotic theory, we propose two bootstrap confidence regions for level sets. The confidence regions can be used to perform tests for anomaly detection and clustering. We also introduce a technique to visualize high dimensional density level sets by combining mode clustering and multidimensional scaling.

stat.ME↗

arXiv2016

Results from the 2014 November 15th multi-chord stellar occultation by the TNO (229762) 2007 UK$_{126}$

We present results derived from the first multi-chord stellar occultation by the trans-Neptunian object (229762) 2007 UK$_{126}$, observed on 2014 November 15. The event was observed by the Research and Education Collaborative Occultation Network (RECON) project and International Occultation Timing Association (IOTA) collaborators throughout the United States. Use of two different data analysis methods obtain a satisfactory fit to seven chords, yelding an elliptical fit to the chords with an equatorial radius of $R=338_{-10} ^{+15}$ km and equivalent radius of $R_{eq}=319_{-7} ^{+14}$ km. A circular fit also gives a radius of $R=324_{-23} ^{+30}$ km. Assuming that the object is a Maclaurin spheroid with indeterminate aspect angle, and using two published absolute magnitudes for the body, we derive possible ranges for geometric albedo between $p_{V}=0.159_{-0.013} ^{+0.007}$ and $p_{R}=0.189_{-0.015}^{+0.009}$, and for the body oblateness between $ε=0.105_{-0.040} ^{+0.050}$ and $ε=0.118_{-0.048} ^{+0.055}$. For a nominal rotational period of 11.05 h, an upper limit for density of $ρ=1740$ kg~m$^{-3}$ is estimated for the body.

astro-ph.EP↗

arXiv2016

Finding Singular Features

We present a method for finding high density, low-dimensional structures in noisy point clouds. These structures are sets with zero Lebesgue measure with respect to the $D$-dimensional ambient space and belong to a $d<D$ dimensional space. We call them "singular features." Hunting for singular features corresponds to finding unexpected or unknown structures hidden in point clouds belonging to $\R^D$. Our method outputs well defined sets of dimensions $d<D$. Unlike spectral clustering, the method works well in the presence of noise. We show how to find singular features by first finding ridges in the estimated density, followed by a filtering step based on the eigenvalues of the Hessian of the density.

stat.ME↗

arXiv2016

Nonparametric modal regression

Modal regression estimates the local modes of the distribution of $Y$ given $X=x$, instead of the mean, as in the usual regression sense, and can hence reveal important structure missed by usual regression methods. We study a simple nonparametric method for modal regression, based on a kernel density estimate (KDE) of the joint distribution of $Y$ and $X$. We derive asymptotic error bounds for this method, and propose techniques for constructing confidence sets and prediction sets. The latter is used to select the smoothing bandwidth of the underlying KDE. The idea behind modal regression is connected to many others, such as mixture regression and density ridge estimation, and we discuss these ties as well.

stat.ME↗

arXiv2016

Minimax Lower Bounds for Linear Independence Testing

Linear independence testing is a fundamental information-theoretic and statistical problem that can be posed as follows: given $n$ points $\{(X_i,Y_i)\}^n_{i=1}$ from a $p+q$ dimensional multivariate distribution where $X_i \in \mathbb{R}^p$ and $Y_i \in\mathbb{R}^q$, determine whether $a^T X$ and $b^T Y$ are uncorrelated for every $a \in \mathbb{R}^p, b\in \mathbb{R}^q$ or not. We give minimax lower bound for this problem (when $p+q,n \to \infty$, $(p+q)/n \leq κ< \infty$, without sparsity assumptions). In summary, our results imply that $n$ must be at least as large as $\sqrt {pq}/\|Σ_{XY}\|_F^2$ for any procedure (test) to have non-trivial power, where $Σ_{XY}$ is the cross-covariance matrix of $X,Y$. We also provide some evidence that the lower bound is tight, by connections to two-sample testing and regression in specific settings.

stat.ML↗

arXiv2015

A Comprehensive Approach to Mode Clustering

Mode clustering is a nonparametric method for clustering that defines clusters using the basins of attraction of a density estimator's modes. We provide several enhancements to mode clustering: (i) a soft variant of cluster assignment, (ii) a measure of connectivity between clusters, (iii) a technique for choosing the bandwidth, (iv) a method for denoising small clusters, and (v) an approach to visualizing the clusters. Combining all these enhancements gives us a complete procedure for clustering in multivariate problems. We also compare mode clustering to other clustering methods in several examples

stat.ME↗