22 results
Search Results
2. Statistical inference for generative adversarial networks and other minimax problems.
- Author
-
Meitz, Mika
- Abstract
This paper studies generative adversarial networks (GANs) from the perspective of statistical inference. A GAN is a popular machine learning method in which the parameters of two neural networks, a generator and a discriminator, are estimated to solve a particular minimax problem. This minimax problem typically has a multitude of solutions and the focus of this paper are the statistical properties of these solutions. We address two key statistical issues for the generator and discriminator network parameters, consistent estimation and confidence sets. We first show that the set of solutions to the sample GAN problem is a (Hausdorff) consistent estimator of the set of solutions to the corresponding population GAN problem. We then devise a computationally intensive procedure to form confidence sets and show that these sets contain the population GAN solutions with the desired coverage probability. Small numerical experiments and a Monte Carlo study illustrate our results and verify our theoretical findings. We also show that our results apply in general minimax problems that may be nonconvex, nonconcave, and have multiple solutions. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
3. Asymptotic inference of the ARMA model with time‐functional variance noises.
- Author
-
Cai, Bibi, Zhu, Enwen, and Ling, Shiqing
- Abstract
This paper studies the autoregressive and moving average (ARMA) model with time‐functional variance (TFV) noises, called the ARMA‐TFV model. We first establish the consistency and asymptotic normality of its least squares estimator (LSE). The Wald tests and portmanteau tests are constructed based on the theory for variable selection and model checking. A simulation study is carried out to assess the performance of our approach in finite samples, and two real examples are given. It should be mentioned that the process generated from the ARMA‐TFV model is not stationary, and the technique in this paper is nonstandard and may provide insights for future research in this area. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
4. Estimation of win, loss probabilities, and win ratio based on right‐censored event data.
- Author
-
Parner, Erik T. and Overgaard, Morten
- Abstract
The win ratio has in the recent decade gained popularity for analyzing prioritized multiple event data in clinical cohort studies, in particular within cardiovascular research. The literature on estimation of the win ratio using censored event data is however sparse. The methods that have been suggested have either an insufficient adjustment of the censoring or by assuming the the win and loss probabilities are proportional over time. The assumption of proportional win and loss probabilities will often in practice not be satisfied. In this paper, we present estimates for the win ratio, and win and loss probabilities, under independent right‐censoring and derive the asymptotic distribution of the estimates. The proposed win ratio estimate does not require the assumption of proportional win and loss probabilities. The small sample properties of the proposed method are studied in a simulation study showing that the variance formula is accurate even for small samples. The method is applied on two data sets. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
5. Nonparametric estimation of densities on the hypersphere using a parametric guide.
- Author
-
Alonso‐Pena, María, Claeskens, Gerda, and Gijbels, Irène
- Abstract
Hyperspherical kernel density estimators (KDE), which use a parametric distribution as a guide, are studied in this paper. The main benefit is that these estimators improve the bias of nonguided kernel density estimators when the parametric guiding distribution is not too far from the true density, while preserving the variance. When using a von Mises‐Fisher density as guide, the proposal performs as well as the classical KDE, even when the guiding model is incorrect, and far from the true distribution. This benefit is particular for the hyperspherical setting given its compact support, and is in contrast to similar methods for real valued data. Moreover, we deal with the important issue of data‐driven selection of the smoothing parameter. Simulations and real data examples illustrate the finite‐sample performance of the proposed method, also in comparison with other recently proposed estimation methods. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
6. Inference for all variants of the multivariate coefficient of variation in factorial designs.
- Author
-
Ditzhaus, Marc and Smaga, Łukasz
- Abstract
The multivariate coefficient of variation (MCV) is an attractive and easy‐to‐interpret effect size for the dispersion in multivariate data. Recently, the first inference methods for the MCV were proposed for general factorial designs. However, the inference methods are primarily derived for one special MCV variant while there are several reasonable proposals. Moreover, when rejecting a global null hypothesis, a more in‐depth analysis is of interest to find the significant contrasts of MCV. This paper concerns extending the nonparametric permutation procedure to the other MCV variants and a max‐type test for post hoc analysis. To improve the small sample performance of the latter, we suggest a novel bootstrap strategy and prove its asymptotic validity. The actual performance of all proposed tests is compared in an extensive simulation study and illustrated by real data analysis. All methods are implemented in the R package GFDmcv, available on CRAN. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
7. Characterization of valid auxiliary functions for representations of extreme value distributions and their max‐domains of attraction.
- Author
-
Seifert, Miriam Isabel
- Subjects
- *
DISTRIBUTION (Probability theory) - Abstract
In this paper we study two important representations for extreme value distributions and their max‐domains of attraction (MDA), namely von Mises representation (vMR) and variation representation (VR), which are convenient ways to gain limit results. Both VR and vMR are defined via so‐called auxiliary functions ψ. Up to now, however, the set of valid auxiliary functions for vMR has neither been characterized completely nor separated from those for VR. We contribute to the current literature by introducing "universal" auxiliary functions which are valid for both VR and vMR representations for the entire MDA distribution families. Then we identify exactly the sets of valid auxiliary functions for both VR and vMR. Moreover, we propose a method for finding appropriate auxiliary functions with analytically simple structure and provide them for several important distributions. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
8. On the expectations of equivariant matrix‐valued functions of Wishart and inverse Wishart matrices.
- Author
-
Hillier, Grant and Kan, Raymond M.
- Subjects
- *
WISHART matrices , *MATRIX inversion , *INVERSE functions , *HOMOGENEOUS spaces , *SYMMETRIC functions - Abstract
Many matrix‐valued functions of an m×m Wishart matrix W, Fk(W), say, are homogeneous of degree k in W, and are equivariant under the conjugate action of the orthogonal group 풪(m), that is, Fk(HWHT)=HFk(W)HT, H∈풪(m). It is easy to see that the expectation of such a function is itself homogeneous of degree k in ∑, the covariance matrix, and are also equivariant under the action of 풪(m) on ∑. The space of such homogeneous, equivariant, matrix‐valued functions is spanned by elements of the type Wrpλ(W), where r∈{0,...,k} and, for each r, λ varies over the partitions of k−r, and pλ(W) denotes the power‐sum symmetric function indexed by λ. In the analogous case where W is replaced by W−1, these elements are replaced by W−rpλ(W−1). In this paper, we derive recurrence relations and analytical expressions for the expectations of such functions. Our results provide highly efficient methods for the computation of all such moments. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
9. Truncated two‐parameter Poisson–Dirichlet approximation for Pitman–Yor process hierarchical models.
- Author
-
Zhang, Junyi and Dassios, Angelos
- Subjects
- *
GIBBS sampling , *APPROXIMATION error , *MARKOV chain Monte Carlo , *COMMON misconceptions - Abstract
In this paper, we construct an approximation to the Pitman–Yor process by truncating its two‐parameter Poisson–Dirichlet representation. The truncation is based on a decreasing sequence of random weights, thus having a lower approximation error compared to the popular truncated stick‐breaking process. We develop an exact simulation algorithm to sample from the approximation process and provide an alternative MCMC algorithm for the parameter regime where the exact simulation algorithm becomes slow. The effectiveness of the simulation algorithms is demonstrated by the estimation of the functionals of a Pitman–Yor process. Then we adapt the approximation process into a Pitman–Yor process mixture model and devise a blocked Gibbs sampler for posterior inference. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
10. Estimation of the adjusted standard‐deviatile for extreme risks.
- Author
-
Chen, Haoyu, Mao, Tiantian, and Yang, Fan
- Subjects
- *
ASYMPTOTIC expansions , *EXTREME value theory , *ASYMPTOTIC normality , *TIME series analysis - Abstract
In this paper, we modify the Bayes risk for the expectile, the so‐called variantile risk measure, to better capture extreme risks. The modified risk measure is called the adjusted standard‐deviatile. First, we derive the asymptotic expansions of the adjusted standard‐deviatile. Next, based on the first‐order asymptotic expansion, we propose two efficient estimation methods for the adjusted standard‐deviatile at intermediate and extreme levels. By using techniques from extreme value theory, the asymptotic normality is proved for both estimators for independent and identically distributed observations and for β‐mixing time series, respectively. Simulations and real data applications are conducted to examine the performance of the proposed estimators. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
11. Density estimation and regression analysis on hyperspheres in the presence of measurement error.
- Author
-
Jeon, Jeong Min and Van Keilegom, Ingrid
- Subjects
- *
REGRESSION analysis , *MEASUREMENT errors , *ASYMPTOTIC normality , *DENSITY , *CONFIDENCE intervals , *DATA analysis , *NONPARAMETRIC estimation - Abstract
This paper studies density estimation and regression analysis with data observed on a general unit hypersphere and contaminated by measurement errors. We establish novel density and regression estimators, and study their asymptotic properties such as the rates of convergence and asymptotic normality. We also provide two types of asymptotic confidence intervals for both density and regression functions. One type is based on the asymptotic normality of their estimators and the other type is based on the empirical likelihood technique. We present practical details on the implementation of our method as well as simulation studies and real data analysis. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
12. Double debiased transfer learning for adaptive Huber regression.
- Author
-
Wang, Ziyuan, Wang, Lei, and Lian, Heng
- Abstract
Through exploiting information from the source data to improve the fit performance on the target data, transfer learning estimations for high‐dimensional linear regression models have drawn much attention recently, but few studies focus on statistical inference and robust learning in the presence of heavy‐tailed/asymmetric errors. Using adaptive Huber regression (AHR) to achieve the bias and robustness tradeoff, in this paper we propose a robust transfer learning algorithm with high‐dimensional covariates, then construct valid confidence intervals and hypothesis tests based on the debiased lasso approach. When the transferable sources are known, a two‐step ℓ1$$ {\ell}_1 $$‐penalized transfer AHR estimator is firstly proposed and the error bounds are established. To correct the biases caused by the lasso penalty, a unified debiasing framework based on the decorrelated score equations is considered to establish asymptotic normality of the debiased lasso transfer AHR estimator. Confidence intervals and hypothesis tests for each component can be constructed. When the transferable sources are unknown, a data‐driven source detection algorithm is proposed with theoretical guarantee. Numerical studies verify the performance of our proposed estimator and confidence intervals, and an application to Genotype‐Tissue Expression data is also presented. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
13. Gradient‐based approach to sufficient dimension reduction with functional or longitudinal covariates.
- Author
-
Huang, Ming‐Yueh and Chan, Kwun Chuen Gary
- Abstract
In this paper, we focus on the sufficient dimension reduction problem in regression analysis with real‐valued response and functional or longitudinal covariates. We propose a new method based on gradients of the conditional distribution function to estimate the sufficient dimension reduction subspace. While existing inverse‐regression‐type methods relies on a linearity condition, our method is based on the gradient of conditional distribution function and its validity only requires smoothness conditions on the population parameters. Practically, the proposed estimator can be obtained by standard algorithm of functional principal component analysis. The proposed method is demonstrated through extensive simulations and two empirical examples. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
14. Cox processes driven by transformed Gaussian processes on linear networks—A review and new contributions.
- Author
-
Møller, Jesper and Rasmussen, Jakob G.
- Abstract
There is a lack of point process models on linear networks. For an arbitrary linear network, we consider new models for a Cox process with an isotropic pair correlation function obtained in various ways by transforming an isotropic Gaussian process which is used for driving the random intensity function of the Cox process. In particular, we introduce three model classes given by log Gaussian, interrupted, and permanental Cox processes on linear networks, and consider for the first time statistical procedures and applications for parametric families of such models. Moreover, we construct new simulation algorithms for Gaussian processes on linear networks and discuss whether the geodesic metric or the resistance metric should be used for the kind of Cox processes studied in this paper. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
15. Mahalanobis balancing: A multivariate perspective on approximate covariate balancing.
- Author
-
Dai, Yimin and Yan, Ying
- Abstract
In the past decade, various exact balancing‐based weighting methods were introduced to the causal inference literature. It eliminates covariate imbalance by imposing balancing constraints in a certain optimization problem, which can nevertheless be infeasible when there is bad overlap between the covariate distributions in the treated and control groups or when the covariates are high dimensional. Recently, approximate balancing was proposed as an alternative balancing framework. It resolves the feasibility issue by using inequality moment constraints instead. However, it can be difficult to select the threshold parameters. Moreover, moment constraints may not fully capture the discrepancy of covariate distributions. In this paper, we propose Mahalanobis balancing to approximately balance covariate distributions from a multivariate perspective. We use a quadratic constraint to control overall imbalance with a single threshold parameter, which can be tuned by a simple selection procedure. We show that the dual problem of Mahalanobis balancing is an ℓ2$$ {\ell}_2 $$ norm‐based regularized regression problem, and establish interesting connection to propensity score models. We derive asymptotic properties, discuss the high‐dimensional scenario, and make extensive numerical comparisons with existing balancing methods. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
16. Nearly unstable integer‐valued ARCH process and unit root testing.
- Author
-
Barreto‐Souza, Wagner and Chan, Ngai Hang
- Subjects
- *
ASYMPTOTIC distribution , *TIME series analysis , *LIMIT theorems , *DEATH rate , *STOCHASTIC integrals , *TOPOLOGY , *ERROR rates , *MONTE Carlo method - Abstract
This paper introduces a Nearly Unstable INteger‐valued AutoRegressive Conditional Heteroscedastic (NU‐INARCH) process for dealing with count time series data. It is proved that a proper normalization of the NU‐INARCH process weakly converges to a Cox–Ingersoll–Ross diffusion in the Skorohod topology. The asymptotic distribution of the conditional least squares estimator of the correlation parameter is established as a functional of certain stochastic integrals. Numerical experiments based on Monte Carlo simulations are provided to verify the behavior of the asymptotic distribution under finite samples. These simulations reveal that the nearly unstable approach provides satisfactory and better results than those based on the stationarity assumption even when the true process is not that close to nonstationarity. A unit root test is proposed and its Type‐I error and power are examined via Monte Carlo simulations. As an illustration, the proposed methodology is applied to the daily number of deaths due to COVID‐19 in the United Kingdom. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
17. Testing the missing at random assumption in generalized linear models in the presence of instrumental variables.
- Author
-
Duan, Rui, Liang, C. Jason, Shaw, Pamela A., Tang, Cheng Yong, and Chen, Yong
- Subjects
- *
INSTRUMENTAL variables (Statistics) , *MISSING data (Statistics) , *DATA analysis , *DECISION making - Abstract
Practical problems with missing data are common, and many methods have been developed concerning the validity and/or efficiency of statistical procedures. On a central focus, there have been longstanding interests on the mechanism governing data missingness, and correctly deciding the appropriate mechanism is crucially relevant for conducting proper practical investigations. In this paper, we present a new hypothesis testing approach for deciding between the conventional notions of missing at random and missing not at random in generalized linear models in the presence of instrumental variables. The foundational idea is to develop appropriate discrepancy measures between estimators whose properties significantly differ only when missing at random does not hold. We show that our testing approach achieves an objective data‐oriented choice between missing at random or not. We demonstrate the feasibility, validity, and efficacy of the new test by theoretical analysis, simulation studies, and a real data analysis. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
18. Communication‐efficient low‐dimensional parameter estimation and inference for high‐dimensional Lp$$ {L}^p $$‐quantile regression.
- Author
-
Gao, Junzhuo and Wang, Lei
- Subjects
- *
QUANTILE regression , *PARAMETER estimation , *CRIME - Abstract
The Lp$$ {L}^p $$‐quantile regression generalizes both quantile regression and expectile regression, and has become popular for its robustness and effectiveness especially when 1
- Published
- 2024
- Full Text
- View/download PDF
19. Locally correct confidence intervals for a binomial proportion: A new criteria for an interval estimator.
- Author
-
Garthwaite, Paul H., Moustafa, Maha W., and Elfadaly, Fadlalla G.
- Subjects
- *
DISTRIBUTION (Probability theory) , *CONFIDENCE intervals - Abstract
Well‐recommended methods of forming "confidence intervals" for a binomial proportion give interval estimates that do not actually meet the definition of a confidence interval, in that their coverages are sometimes lower than the nominal confidence level. The methods are favoured because their intervals have a shorter average length than the Clopper–Pearson (gold‐standard) method, whose intervals really are confidence intervals. As the definition of a confidence interval is not being adhered to, another criterion for forming interval estimates for a binomial proportion is needed. In this paper, we suggest a new criterion for forming one‐sided intervals and equal‐tail two‐sided intervals. Methods which meet the criterion are said to yield locally correct confidence intervals. We propose a method that yields such intervals and prove that its intervals have a shorter average length than those of any other method that meets the criterion. Compared with the Clopper–Pearson method, the proposed method gives intervals with an appreciably smaller average length. For confidence levels of practical interest, the mid‐p method also satisfies the new criterion and has its own optimality property. It gives locally correct confidence intervals that are only slightly wider than those of the new method. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
20. Estimating absorption time distributions of general Markov jump processes.
- Author
-
Ahmad, Jamaal, Bladt, Martin, and Bladt, Mogens
- Subjects
- *
JUMP processes , *MARKOV processes , *MATRIX analytic methods , *EXPECTATION-maximization algorithms , *LEVY processes , *MAXIMUM likelihood statistics , *MATRIX functions - Abstract
The estimation of absorption time distributions of Markov jump processes is an important task in various branches of statistics and applied probability. While the time‐homogeneous case is classic, the time‐inhomogeneous case has recently received increased attention due to its added flexibility and advances in computational power. However, commuting sub‐intensity matrices are assumed, which in various cases limits the parsimonious properties of the resulting representation. This paper develops the theory required to solve the general case through maximum likelihood estimation, and in particular, using the expectation‐maximization algorithm. A reduction to a piecewise constant intensity matrix function is proposed in order to provide succinct representations, where a parametric linear model binds the intensities together. Practical aspects are discussed and illustrated through the estimation of notoriously demanding theoretical distributions and real data, from the perspective of matrix analytic methods. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
21. Locally adaptive Bayesian isotonic regression using half shrinkage priors.
- Author
-
Okano, Ryo, Hamura, Yasuyuki, Irie, Kaoru, and Sugasawa, Shonosuke
- Subjects
- *
ISOTONIC regression , *GIBBS sampling , *RANDOM variables - Abstract
Isotonic regression or monotone function estimation is a problem of estimating function values under monotonicity constraints, which appears naturally in many scientific fields. This paper proposes a new Bayesian method with global–local shrinkage priors for estimating monotone function values. Specifically, we introduce half shrinkage priors for positive valued random variables and assign them for the first‐order differences of function values. We also develop fast and simple Gibbs sampling algorithms for full posterior analysis. By incorporating advanced shrinkage priors, the proposed method is adaptive to local abrupt changes or jumps in target functions. We show this adaptive property theoretically by proving that the posterior mean estimators are robust to large differences and that asymptotic risk for unchanged points can be improved. Finally, we demonstrate the proposed methods through simulations and applications to a real data set. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
22. Extrapolation estimation for nonparametric regression with measurement error.
- Author
-
Song, Weixing, Ayub, Kanwal, and Shi, Jianhong
- Subjects
- *
MEASUREMENT errors , *NONPARAMETRIC estimation , *EXTRAPOLATION , *CONDITIONAL expectations , *REGRESSION analysis - Abstract
For the nonparametric regression models with covariates contaminated with the normal measurement errors, this paper proposes an extrapolation algorithm to estimate the regression functions. By applying the conditional expectation directly to the kernel‐weighted least squares of the deviations between the local linear approximation and the observed responses, the proposed algorithm successfully bypasses the simulation step in the classical simulation extrapolation, thus significantly reducing the computational time. It is noted that the proposed method also provides an exact form of the extrapolation function, although the extrapolation estimate generally cannot be obtained by simply setting the extrapolation variable to negative one in the fitted extrapolation function, if the bandwidth is less than the SD of the measurement error. Large sample properties of the proposed estimation procedure are discussed, as well as simulation studies and a real data example being conducted to illustrate its applications. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.