Filters








756 Hits in 7.0 sec

Inference in High-Dimensional Panel Models With an Application to Gun Control

Alexandre Belloni, Victor Chernozhukov, Christian Hansen, Damian Kozbur
2016 Journal of business & economic statistics  
We consider estimation and inference in panel data models with additive unobserved individual specific heterogeneity in a high dimensional setting.  ...  endogenous variables in panel data instrumental variables models with fixed effects and many instruments.  ...  standard errors and heteroscedastic standard errors for each estimator.  ... 
doi:10.1080/07350015.2015.1102733 fatcat:x7fq5camg5bmdnengdllccs6fi

Minimax wavelet estimation for multisample heteroscedastic non-parametric regression [article]

Madison Giacofc and Sophie Lambert-Lacroix and Franck Picard
2015 arXiv   pre-print
This framework allows us to model curves that can exhibit strong irregularities, such as peaks or jumps for instance.  ...  The lower bound for the L_2 minimax risk is provided, as well as the upper bound of the minimax rate, that is derived by constructing a wavelet estimator for the functional fixed effect.  ...  We are grateful to Anatoli Juditsky for constructive and fruitful discussions.  ... 
arXiv:1511.04556v1 fatcat:2mfmi4gpt5bwfhdilhecgij5ce

Functional Regression [article]

Jeffrey S. Morris
2014 arXiv   pre-print
The primary focus is on modeling and methodology, highlighting the modeling structures that have been developed and the various regularization approaches employed.  ...  For each, the role of replication and regularization will be discussed and the methodological development described in a roughly chronological manner, at times deviating from the historical timeline to  ...  proposed a semi-parametric model that adds parametric fixed effects to a model with a nonparametric mean function, using periodic spline bases for the mean curve, regularizing by roughness penalty.  ... 
arXiv:1406.4068v1 fatcat:uvbzr45s75etpl7faj6y7wkz6i

Maximum likelihood multiple subspace projections for hidden Markov models

M.J.F. Gales
2002 IEEE Transactions on Speech and Audio Processing  
Efficient estimation formulae for the model parameters for both schemes are derived. In addition, the computational cost for their use during recognition are given.  ...  For both schemes, the problem of handling likelihood consistency between the various subspaces is dealt with by viewing the projection schemes within a maximum likelihood framework.  ...  Recently the use of maximum likelihood (ML) estimation has been proposed for generating a linear subspace projection 1 with class labeled data, heteroscedastic LDA (HLDA) [3] .  ... 
doi:10.1109/89.985541 fatcat:433m7oubm5h3pbroky2tlw624i

Adaptive Estimation in Multivariate Response Regression with Hidden Variables [article]

Xin Bing and Yang Ning and Yaosheng Xu
2021 arXiv   pre-print
The model identifiability, parameter estimation and statistical guarantees are further extended to the setting with heteroscedastic errors.  ...  In the last step, we remove the effect of hidden variable by projecting Y onto the complement of the estimated row space of B^*. Non-asymptotic error bounds of our final estimator are established.  ...  To conclude this section, we compare the estimation errors of P B * and the PCA-based estimator P B * in Section 2.2.2 in the presence of heteroscedasticity.  ... 
arXiv:2003.13844v2 fatcat:dnrk4afde5a4hiuhudkm36raqm

Hybrid Orthogonal Projection and Estimation (HOPE): A New Framework to Probe and Learn Neural Networks [article]

Shiliang Zhang, Hui Jiang
2015 arXiv   pre-print
In this paper, we propose a novel model for high-dimensional data, called the Hybrid Orthogonal Projection and Estimation (HOPE) model, which combines a linear orthogonal projection and a finite mixture  ...  The HOPE model itself can be learned unsupervised from unlabelled data based on the maximum likelihood estimation as well as discriminatively from labelled data.  ...  In semi-unsupervised learning, the HOPE model is learned based on the maximum likelihood estimation and the upper deep NN is learned supervised.  ... 
arXiv:1502.00702v2 fatcat:uigv3saktfc4zgm77h44d4rrry

A spectral series approach to high-dimensional nonparametric regression

Ann B. Lee, Rafael Izbicki
2016 Electronic Journal of Statistics  
In this work, we present an orthogonal series estimator for predictors that are complex aggregate objects, such as natural images, galaxy spectra, trajectories, and movies.  ...  We provide theoretical guarantees for a radial kernel with varying bandwidth, and we relate smoothness of the regression function with respect to P to sparsity in the eigenbasis.  ...  Coifman, Stéphane Lafon, and Larry Wasserman for the original discussions that led to this work.  ... 
doi:10.1214/16-ejs1112 fatcat:vsuh2clwjfaoxniawzdqvgoloy

Parametric emulation and inference in computationally expensive integrated urban water quality simulators

Antonio M. Moreno-Rodenas, Jeroen G. Langeveld, Francois H. L. R. Clemens
2019 Environmental science and pollution research international  
The complexity of these integrated catchment models grows fast, leading to potentially over-parameterised and computationally expensive models.  ...  The effect of different likelihood assumptions (e.g. heteroscedasticity, normality and autocorrelation) during the inference of dissolved oxygen processes is also discussed.  ...  model parameters. σ 1 and σ 2 are hyperparameters of the selected error generation process (heteroscedastic, independent Gaussian).  ... 
doi:10.1007/s11356-019-05620-1 pmid:31273657 fatcat:dlxzxvz6bfczxptvetfhepxns4

Improved Covariance Matrix Estimation for Portfolio Risk Measurement: A Review

Ruili Sun, Tiefeng Ma, Shuangzhe Liu, Milind Sathye
2019 Journal of Risk and Financial Management  
Moment-based risk measurement includes time-varying covariance matrix and shrinkage estimation, while moment-based and quantile-based risk measurement includes semi-variance, VaR and CVaR.  ...  In the portfolio selection model part, we cover three models: mean-variance model, global minimum variance (GMV) model and factor model.  ...  Acknowledgments: The authors would like to thank the editor and referees for the opportunity and their constructive comments which led to an improved version of the manuscript.  ... 
doi:10.3390/jrfm12010048 fatcat:7vancjldk5chrfabmoammhnqhi

Non-Negativity of a Quadratic form with Applications to Panel Data Estimation, Forecasting and Optimization

Bhimasankaram Pochiraju, Sridhar Seshadri, Dimitrios D. Thomakos, Konstantinos Nikolopoulos
2020 Stats  
We show that the test can be performed if the estimated error variances in the fixed and random effects models satisfy a specific inequality.  ...  For a symmetric matrix B, we determine the class of Q such that Q t BQ is non-negative definite and apply it to panel data estimation and forecasting: the Hausman test for testing the endogeneity of the  ...  coefficient estimators in the fixed effects and random effects models (with homoscedastic structures for the error in the fixed effects model and for the random error and also for the random effects in  ... 
doi:10.3390/stats3030015 fatcat:ekx2uv73prd6li4ilnkmt7svpa

Functional CAR Models for Large Spatially Correlated Functional Datasets

Lin Zhang, Veerabhadran Baladandayuthapani, Hongxiao Zhu, Keith A. Baggerly, Tadeusz Majewski, Bogdan A. Czerniak, Jeffrey S. Morris
2016 Journal of the American Statistical Association  
We show theoretically that our construction leads to a CAR model at each functional location, with spatial covariance parameters varying and borrowing strength across the functional domain.  ...  Our model performs functional response regression while accounting for spatial correlations with potentially nonseparable and nonstationary covariance structure, in both the space and functional domains  ...  In addition, the functional CAR model is generalizable for functional analysis using different basis functions, orthogonal or non-orthogonal, based on the data characteristics and analytic purpose, and  ... 
doi:10.1080/01621459.2015.1042581 pmid:28018013 pmcid:PMC5176110 fatcat:frredrgivfaqrh5arz2tokw5fi

HIGH-FIDELITY RADIO ASTRONOMICAL POLARIMETRY USING A MILLISECOND PULSAR AS A POLARIZED REFERENCE SOURCE

W. van Straten
2013 Astrophysical Journal Supplement Series  
Application of the new technique followed by arrival time estimation using matrix template matching yields post-fit residuals with an uncertainty-weighted standard deviation of 880 ns, two times smaller  ...  The precision achieved by this experiment yields the first significant measurements of the secular variation of the projected semi-major axis, the precession of periastron, and the Shapiro delay; it also  ...  The best-fit model parameters also include the first significant detections of the precession of periastronω and the secular variation of the projected semi-major axis of the orbit,ẋ.  ... 
doi:10.1088/0067-0049/204/1/13 fatcat:r4dtzyf7xrbepcikkszdz2qpju

Valid Post-Selection and Post-Regularization Inference: An Elementary, General Approach

Victor Chernozhukov, Christian Hansen, Martin Spindler
2015 Annual Review of Economics  
conditions on the smoothness of M and the quality of the estimator η̂, guarantees that inference on for the main parameter α based on testing or point estimation methods discussed below will be regular  ...  Simple, readily verifiable sufficient conditions are provided for a class of affine-quadratic models.  ...  With the larger set of variables, our post-model-selection estimator of the price coefficient is -.221 with an estimated standard error of .015 compared to the OLS estimate of -.099 with an estimated standard  ... 
doi:10.1146/annurev-economics-012315-015826 fatcat:e6jn3aaffff47m7gx2a2akwtuq

Valid Post-Selection and Post-Regularization Inference: An Elementary, General Approach

Victor Chernozhukov, Christian Hansen, Martin Spindler
2014 Social Science Research Network  
Our analysis provides a set of high-level conditions under which inference for the low-dimensional parameter based on testing or point estimation methods will be regular despite selection or regularization  ...  We conclude with a review of other developments in postselection inference and note that many of the developments can be viewed as special cases of the general encompassing framework of orthogonal estimating  ...  With the larger set of variables, our post-model-selection estimator of the price coefficient is -.221 with an estimated standard error of .015 compared to the OLS estimate of -.099 with an estimated standard  ... 
doi:10.2139/ssrn.2566887 fatcat:wniu733gezdnnfcsbqunv256uy

Spatial Interactions in Hedonic Pricing Models: The Urban Housing Market of Aveiro, Portugal

Arnab Bhattacharjee, Eduardo Castro, João Marques
2012 Spatial Economic Analysis  
Specifically, based on an a priori definition of housing submarkets and the assumption of a multifactor model, we develop maximum likelihood methodology to estimate hedonic models that facilitate understanding  ...  The methodology, based on statistical orthogonal factor analysis, is applied to the urban housing market of Aveiro, Portugal at two different spatial scales.  ...  model in semi-log form.  ... 
doi:10.1080/17421772.2011.647058 fatcat:bdm2zkoa4nhntkvei6yijorsya
« Previous Showing results 1 — 15 out of 756 results