arXiv subjects

Peter Hall

Publications and source records attributed to Peter Hall.

At least 37 records · Page 2Linked to original sources

arXiv2014

Understanding Instrumental Stokes Leakage in Murchison Widefield Array Polarimetry

This paper offers an electromagnetic, more specifically array theory, perspective on understanding strong instrumental polarization effects for planar low-frequency "aperture arrays" with the Murchison Widefield Array (MWA) as an example. A long-standing issue that has been seen here is significant instrumental Stokes leakage after calibration, particularly in Stokes Q at high frequencies. A simple model that accounts for inter-element mutual coupling is presented which explains the prominence of Q leakage seen when the array is scanned away from zenith in the principal planes. On these planes, the model predicts current imbalance in the X (E-W) and Y (N-S) dipoles and hence the Q leakage. Although helpful in concept, we find that this model is inadequate to explain the full details of the observation data. This finding motivates further experimentation with more rigorous models that account for both mutual coupling and embedded element patterns. Two more rigorous models are discussed: the "full" and "average" embedded element patterns. The viability of the "full" model is demonstrated by simulating current MWA practice of using a Hertzian dipole model as a Jones matrix estimate. We find that these results replicate the observed Q leakage to approximately 2 to 5%. Finally, we offer more direct indication for the level of improvement expected from upgrading the Jones matrix estimate with more rigorous models. Using the "average" embedded pattern as an estimate for the "full" model, we find that Q leakage of a few percent is achievable.

astro-ph.IM↗

arXiv2014

Double-bootstrap methods that use a single double-bootstrap simulation

We show that, when the double bootstrap is used to improve performance of bootstrap methods for bias correction, techniques based on using a single double-bootstrap sample for each single-bootstrap sample can be particularly effective. In particular, they produce third-order accuracy for much less computational expense than is required by conventional double-bootstrap methods. However, this improved level of performance is not available for the single double-bootstrap methods that have been suggested to construct confidence intervals or distribution estimators.

math.ST↗

arXiv2014

Feature selection when there are many influential features

Recent discussion of the success of feature selection methods has argued that focusing on a relatively small number of features has been counterproductive. Instead, it is suggested, the number of significant features can be in the thousands or tens of thousands, rather than (as is commonly supposed at present) approximately in the range from five to fifty. This change, in orders of magnitude, in the number of influential features, necessitates alterations to the way in which we choose features and to the manner in which the success of feature selection is assessed. In this paper, we suggest a general approach that is suited to cases where the number of relevant features is very large, and we consider particular versions of the approach in detail. We propose ways of measuring performance, and we study both theoretical and numerical properties of the proposed methodology.

math.ST↗

arXiv2014

Truncated Linear Models for Functional Data

A conventional linear model for functional data involves expressing a response variable $Y$ in terms of the explanatory function $X(t)$, via the model: $Y=a+\int_I b(t)X(t)dt+\hbox{error}$, where $a$ is a scalar, $b$ is an unknown function and $I=[0, α]$ is a compact interval. However, in some problems the support of $b$ or $X$, $I_1$ say, is a proper and unknown subset of $I$, and is a quantity of particular practical interest. In this paper, motivated by a real-data example involving particulate emissions, we develop methods for estimating $I_1$. We give particular emphasis to the case $I_1=[0,θ]$, where $θ\in(0,α]$, and suggest two methods for estimating $a$, $b$ and $θ$ jointly; we introduce techniques for selecting tuning parameters; and we explore properties of our methodology using both simulation and the real-data example mentioned above. Additionally, we derive theoretical properties of the methodology, and discuss implications of the theory. Our theoretical arguments give particular emphasis to the problem of identifiability.

stat.ME↗

arXiv2014

An Algorithm for Nonlinear, Nonparametric Model Choice and Prediction

We introduce an algorithm which, in the context of nonlinear regression on vector-valued explanatory variables, chooses those combinations of vector components that provide best prediction. The algorithm devotes particular attention to components that might be of relatively little predictive value by themselves, and so might be ignored by more conventional methodology for model choice, but which, in combination with other difficult-to-find components, can be particularly beneficial for prediction. Additionally the algorithm avoids choosing vector components that become redundant once appropriate combinations of other, more relevant components are selected. It is suitable for very high dimensional problems, where it keeps computational labour in check by using a novel sequential argument, and also for more conventional prediction problems, where dimension is relatively low. We explore properties of the algorithm using both theoretical and numerical arguments.

stat.ME↗

arXiv2013

Unexpected properties of bandwidth choice when smoothing discrete data for constructing a functional data classifier

The data functions that are studied in the course of functional data analysis are assembled from discrete data, and the level of smoothing that is used is generally that which is appropriate for accurate approximation of the conceptually smooth functions that were not actually observed. Existing literature shows that this approach is effective, and even optimal, when using functional data methods for prediction or hypothesis testing. However, in the present paper we show that this approach is not effective in classification problems. There a useful rule of thumb is that undersmoothing is often desirable, but there are several surprising qualifications to that approach. First, the effect of smoothing the training data can be more significant than that of smoothing the new data set to be classified; second, undersmoothing is not always the right approach, and in fact in some cases using a relatively large bandwidth can be more effective; and third, these perverse results are the consequence of very unusual properties of error rates, expressed as functions of smoothing parameters. For example, the orders of magnitude of optimal smoothing parameter choices depend on the signs and sizes of terms in an expansion of error rate, and those signs and sizes can vary dramatically from one setting to another, even for the same classifier.

math.ST↗

arXiv2013

Normal approximation and smoothness for sums of means of lattice-valued random variables

Motivated by a problem arising when analysing data from quarantine searches, we explore properties of distributions of sums of independent means of independent lattice-valued random variables. The aim is to determine the extent to which approximations to those sums require continuity corrections. We show that, in cases where there are only two different means, the main effects of distribution smoothness can be understood in terms of the ratio $ρ_{12}=(e_2n_1)/(e_1n_2)$, where $e_1$ and $e_2$ are the respective maximal lattice edge widths of the two populations, and $n_1$ and $n_2$ are the respective sample sizes used to compute the means. If $ρ_{12}$ converges to an irrational number, or converges sufficiently slowly to a rational number; and in a number of other cases too, for example those where $ρ_{12}$ does not converge; the effects of the discontinuity of lattice distributions are of smaller order than the effects of skewness. However, in other instances, for example where $ρ_{12}$ converges relatively quickly to a rational number, the effects of discontinuity and skewness are of the same size. We also treat higher-order properties, arguing that cases where $ρ_{12}$ converges to an algebraic irrational number can be less prone to suffer the effects of discontinuity than cases where the limiting irrational is transcendental. These results are extended to the case of three or more different means, and also to problems where distributions are estimated using the bootstrap. The results have practical interpretation in terms of the accuracy of inference for, among other quantities, the sum or difference of binomial proportions.

math.ST↗

arXiv2013

A simple bootstrap method for constructing nonparametric confidence bands for functions

Standard approaches to constructing nonparametric confidence bands for functions are frustrated by the impact of bias, which generally is not estimated consistently when using the bootstrap and conventionally smoothed function estimators. To overcome this problem it is common practice to either undersmooth, so as to reduce the impact of bias, or oversmooth, and thereby introduce an explicit or implicit bias estimator. However, these approaches, and others based on nonstandard smoothing methods, complicate the process of inference, for example, by requiring the choice of new, unconventional smoothing parameters and, in the case of undersmoothing, producing relatively wide bands. In this paper we suggest a new approach, which exploits to our advantage one of the difficulties that, in the past, has prevented an attractive solution to the problem - the fact that the standard bootstrap bias estimator suffers from relatively high-frequency stochastic error. The high frequency, together with a technique based on quantiles, can be exploited to dampen down the stochastic error term, leading to relatively narrow, simple-to-construct confidence bands.

math.ST↗

arXiv2012

Single and multiple index functional regression models with nonparametric link

Fully nonparametric methods for regression from functional data have poor accuracy from a statistical viewpoint, reflecting the fact that their convergence rates are slower than nonparametric rates for the estimation of high-dimensional functions. This difficulty has led to an emphasis on the so-called functional linear model, which is much more flexible than common linear models in finite dimension, but nevertheless imposes structural constraints on the relationship between predictors and responses. Recent advances have extended the linear approach by using it in conjunction with link functions, and by considering multiple indices, but the flexibility of this technique is still limited. For example, the link may be modeled parametrically or on a grid only, or may be constrained by an assumption such as monotonicity; multiple indices have been modeled by making finite-dimensional assumptions. In this paper we introduce a new technique for estimating the link function nonparametrically, and we suggest an approach to multi-index modeling using adaptively defined linear projections of functional data. We show that our methods enable prediction with polynomial convergence rates. The finite sample performance of our methods is studied in simulations, and is illustrated by an application to a functional regression problem.

math.ST↗

arXiv2012

Methodology and theory for partial least squares applied to functional data

The partial least squares procedure was originally developed to estimate the slope parameter in multivariate parametric models. More recently it has gained popularity in the functional data literature. There, the partial least squares estimator of slope is either used to construct linear predictive models, or as a tool to project the data onto a one-dimensional quantity that is employed for further statistical analysis. Although the partial least squares approach is often viewed as an attractive alternative to projections onto the principal component basis, its properties are less well known than those of the latter, mainly because of its iterative nature. We develop an explicit formulation of partial least squares for functional data, which leads to insightful results and motivates new theory, demonstrating consistency and establishing convergence rates.

math.ST↗

arXiv2012

Nonparametric regression with homogeneous group testing data

We introduce new nonparametric predictors for homogeneous pooled data in the context of group testing for rare abnormalities and show that they achieve optimal rates of convergence. In particular, when the level of pooling is moderate, then despite the cost savings, the method enjoys the same convergence rate as in the case of no pooling. In the setting of "over-pooling" the convergence rate differs from that of an optimal estimator by no more than a logarithmic factor. Our approach improves on the random-pooling nonparametric predictor, which is currently the only nonparametric method available, unless there is no pooling, in which case the two approaches are identical.

math.ST↗

arXiv2012

Asymptotic normality and valid inference for Gaussian variational approximation

We derive the precise asymptotic distributional behavior of Gaussian variational approximate estimators of the parameters in a single-predictor Poisson mixed model. These results are the deepest yet obtained concerning the statistical properties of a variational approximation method. Moreover, they give rise to asymptotically valid statistical inference. A simulation study demonstrates that Gaussian variational approximate confidence intervals possess good to excellent coverage properties, and have a similar precision to their exact likelihood counterparts.

math.ST↗

arXiv2010

Modeling the variability of rankings

For better or for worse, rankings of institutions, such as universities, schools and hospitals, play an important role today in conveying information about relative performance. They inform policy decisions and budgets, and are often reported in the media. While overall rankings can vary markedly over relatively short time periods, it is not unusual to find that the ranks of a small number of "highly performing" institutions remain fixed, even when the data on which the rankings are based are extensively revised, and even when a large number of new institutions are added to the competition. In the present paper, we endeavor to model this phenomenon. In particular, we interpret as a random variable the value of the attribute on which the ranking should ideally be based. More precisely, if $p$ items are to be ranked then the true, but unobserved, attributes are taken to be values of $p$ independent and identically distributed variates. However, each attribute value is observed only with noise, and via a sample of size roughly equal to $n$, say. These noisy approximations to the true attributes are the quantities that are actually ranked. We show that, if the distribution of the true attributes is light-tailed (e.g., normal or exponential) then the number of institutions whose ranking is correct, even after recalculation using new data and even after many new institutions are added, is essentially fixed. Formally, $p$ is taken to be of order $n^C$ for any fixed $C>0$, and the number of institutions whose ranking is reliable depends very little on $p$.

math.ST↗

arXiv2010

Strong approximations of level exceedences related to multiple hypothesis testing

Particularly in genomics, but also in other fields, it has become commonplace to undertake highly multiple Student's $t$-tests based on relatively small sample sizes. The literature on this topic is continually expanding, but the main approaches used to control the family-wise error rate and false discovery rate are still based on the assumption that the tests are independent. The independence condition is known to be false at the level of the joint distributions of the test statistics, but that does not necessarily mean, for the small significance levels involved in highly multiple hypothesis testing, that the assumption leads to major errors. In this paper, we give conditions under which the assumption of independence is valid. Specifically, we derive a strong approximation that closely links the level exceedences of a dependent ``studentized process'' to those of a process of independent random variables. Via this connection, it can be seen that in high-dimensional, low sample-size cases, provided the sample size diverges faster than the logarithm of the number of tests, the assumption of independent $t$-tests is often justified.

math.ST↗

arXiv2010

Innovated higher criticism for detecting sparse signals in correlated noise

Higher criticism is a method for detecting signals that are both sparse and weak. Although first proposed in cases where the noise variables are independent, higher criticism also has reasonable performance in settings where those variables are correlated. In this paper we show that, by exploiting the nature of the correlation, performance can be improved by using a modified approach which exploits the potential advantages that correlation has to offer. Indeed, it turns out that the case of independent noise is the most difficult of all, from a statistical viewpoint, and that more accurate signal detection (for a given level of signal sparsity and strength) can be obtained when correlation is present. We characterize the advantages of correlation by showing how to incorporate them into the definition of an optimal detection boundary. The boundary has particularly attractive properties when correlation decays at a polynomial rate or the correlation matrix is Toeplitz.

math.ST↗

arXiv2010

Local polynomial regression and variable selection

We propose a method for incorporating variable selection into local polynomial regression. This can improve the accuracy of the regression by extending the bandwidth in directions corresponding to those variables judged to be are unimportant. It also increases our understanding of the dataset by highlighting areas where these variables are redundant. The approach has the potential to effect complete variable removal as well as perform partial removal when a variable redundancy applies only to particular regions of the data. We define a nonparametric oracle property and show that this is more than satisfied by our approach under asymptotic analysis. The usefulness of the method is demonstrated through simulated and real data numerical examples.

math.ST↗

arXiv2010

Kernel methods and minimum contrast estimators for empirical deconvolution

We survey classical kernel methods for providing nonparametric solutions to problems involving measurement error. In particular we outline kernel-based methodology in this setting, and discuss its basic properties. Then we point to close connections that exist between kernel methods and much newer approaches based on minimum contrast techniques. The connections are through use of the sinc kernel for kernel-based inference. This `infinite order' kernel is not often used explicitly for kernel-based deconvolution, although it has received attention in more conventional problems where measurement error is not an issue. We show that in a comparison between kernel methods for density deconvolution, and their counterparts based on minimum contrast, the two approaches give identical results on a grid which becomes increasingly fine as the bandwidth decreases. In consequence, the main numerical differences between these two techniques are arguably the result of different approaches to choosing smoothing parameters.

stat.ME↗

arXiv2010

Defining probability density for a distribution of random functions

The notion of probability density for a random function is not as straightforward as in finite-dimensional cases. While a probability density function generally does not exist for functional data, we show that it is possible to develop the notion of density when functional data are considered in the space determined by the eigenfunctions of principal component analysis. This leads to a transparent and meaningful surrogate for density defined in terms of the average value of the logarithms of the densities of the distributions of principal components for a given dimension. This density approximation is estimable readily from data. It accurately represents, in a monotone way, key features of small-ball approximations to density. Our results on estimators of the densities of principal component scores are also of independent interest; they reveal interesting shape differences that have not previously been considered. The statistical implications of these results and properties are identified and discussed, and practical ramifications are illustrated in numerical work.

math.ST↗