Journal of the American Statistical Association

最新文章

• Inside-Out Cross-Covariance for Spatial Multivariate Data AI翻译
  作者：Peruzzi, Michele
  作者单位：University of Michigan System; University of Michigan
  摘要：As the spatial features of multivariate data are increasingly central in researchers' applied problems, there is a growing demand for novel spatially aware methods that are flexible, easily interpretable, and scalable to large data. We develop inside-out cross-covariance (IOX) models for multivariate spatial likelihood-based inference. IOX leads to valid cross-covariance matrix functions which we interpret as inducing spatial dependence on independent replicates of a correlated random vector. ...
• Theory for Identification and Inference with Synthetic Controls: A Proximal Causal Inference Framework AI翻译
  作者：Shi, Xu; Li, Kendrick Qijun; Yu, Myeonghun; Miao, Wang; Kuchibhotla, Arun Kumar; Hu, Mengtong; Tchetgen Tchetgen, Eric
  作者单位：University of Michigan System; University of Michigan; St Jude Children's Research Hospital; Peking University; Carnegie Mellon University; University of Pennsylvania
  摘要：Synthetic control (SC) methods are commonly used to estimate the treatment effect on a single treated unit in panel data settings. An SC is a weighted average of control units built to match the treated unit, with weights typically estimated by regressing pretreatment outcomes and measured covariates of the treated unit to those of the control units. However, the classical SC method was primarily proposed for empirical settings where a good pretreatment fit is attainable. In this article, we i...
• Trans-Glasso: A Transfer Learning Approach to Precision Matrix Estimation AI翻译
  作者：Zhao, Boxin; Ma, Cong; Kolar, Mladen
  作者单位：University of Chicago; University of Chicago; University of Southern California; Mohamed bin Zayed University of Artificial Intelligence MBZUAI
  摘要：Precision matrix estimation is essential in various fields, yet it is challenging when samples for the target study are limited. Transfer learning can enhance estimation accuracy by leveraging data from related source studies. We propose Trans-Glasso, a two-step transfer learning method for precision matrix estimation. First, we obtain initial estimators using a multi-task learning objective that captures both shared and unique features across studies. Then, we refine these estimators through ...
• Estimating Multiple Structural Breaks in Large Panels With Unobserved Heterogeneity AI翻译
  作者：Huang, Wenxin; Sun, Yucheng; Wang, Yiru
  作者单位：Shanghai Jiao Tong University; Capital University of Economics & Business; Pennsylvania Commonwealth System of Higher Education (PCSHE); University of Pittsburgh
  摘要：This article proposes a novel algorithm to identify potentially multiple breaks in linear panel data models, while the slope coefficients can have individual heterogeneity and the cross-sectional dimension is allowed to diverge jointly with the time span. The algorithm adopts a shrinkage penalty to detect breaks, and delivers a consistent estimator for the number of breaks and fraction-consistent estimators for break dates. In the presence of latent grouped heterogeneity, we employ the classif...
• Spatial Variation on Multiple Scales in Line Transect Data; the Case of Antarctic Fin Whales AI翻译
  作者：Breivik, Olav Nikolai; Skaug, Hans J.; Jullum, Martin; Biuw, Martin
  作者单位：University of Bergen; Institute of Marine Research - Norway; Institute of Marine Research - Norway
  摘要：Line transect sampling is a widely used survey method for estimating animal density or abundance. We present a novel model for such data that allows for spatial variation in animal density at two scales: a long scale, representing trends caused by for instance climatic or terrain gradients, and a short scale, representing abrupt shifts due to local effects such as prey patchiness. The long-range variation is modeled as a latent Gaussian random field, while the abrupt changes are modeled as a t...
• Dirichlet-Laplace priors for optimal shrinkage (vol 110, pg1479-1490, 2014 ) AI翻译
  作者：Gruber, Luis; Kastner, Gregor; Bhattacharya, Anirban; Pati, Debdeep; Pillai, Natesh; Dunson, David
  作者单位：University of Klagenfurt; Texas A&M University System; Texas A&M University College Station; University of Wisconsin System; University of Wisconsin Madison; Harvard University; Duke University
  摘要：Bhattacharya et al. introduce a novel prior, the Dirichlet-Laplace (DL) prior, and propose a Markov chain Monte Carlo (MCMC) method to simulate posterior draws under this prior in a conditionally Gaussian setting. The original algorithm samples from conditional distributions in the wrong order, that is, it does not correctly sample from the joint posterior distribution of all latent variables. This note details the issue and provides two simple solutions: A correction to the original algorithm...
• Data-Driven Tuning Parameter Selection for High-Dimensional Vector Autoregressions AI翻译
  作者：Kock, Anders B.; Pedersen, Rasmus S.; Sorensen, Jesper R. -V.
  作者单位：University of Oxford; University of Copenhagen; Danish Finance Institute
  摘要：Lasso-type estimators are routinely used to estimate high-dimensional time series models. The theoretical guarantees established for these estimators typically require the penalty level to be chosen in a suitable fashion often depending on unknown population quantities. Furthermore, the resulting estimates and the number of variables retained in the model depend crucially on the chosen penalty level. However, there is currently no theoretically founded guidance for this choice in the context o...
• Euclidean Mirrors and Dynamics in Network Time Series AI翻译
  作者：Athreya, Avanti; Lubberts, Zachary; Park, Youngser; Priebe, Carey
  作者单位：Johns Hopkins University; University of Virginia; Johns Hopkins University
  摘要：Analyzing changes in network evolution is central to statistical network inference. We consider a dynamic network model in which each node has an associated time-varying low-dimensional latent vector of feature data, and connection probabilities are functions of these vectors. Under mild assumptions, the evolution of latent vectors exhibits low-dimensional manifold structure under a suitable distance. This distance can be approximated by a measure of separation between the observed networks th...
• Data Fission and Sampling Designs: A Discussion AI翻译
  作者：Chan, Kwun Chuen Gary; Prentice, Ross L.; Yuan, Zhenman
  作者单位：University of Washington; University of Washington Seattle; Fred Hutchinson Cancer Center; University of Washington; University of Washington Seattle
• Bayesian Transfer Learning for Enhanced Estimation and Inference AI翻译
  作者：Lai, Daoyuan; Madrid Padilla, Oscar Hernan; Li, Xiang; Gu, Tian
  作者单位：Columbia University; Columbia University; University of California System; University of California Los Angeles; University of Texas System; UTMD Anderson Cancer Center
  摘要：Transfer learning enhances model performance in a target population with limited samples by leveraging knowledge from related studies. While many works focus on improving predictive performance, challenges persist in statistical inference. Bayesian approaches naturally provide uncertainty quantification for parameter estimates; however, existing Bayesian transfer learning methods are typically limited to single-source scenarios or require individual-level data. We introduce TRansfer leArning v...