Heteroscedasticity-aware residuals-based contextual stochastic optimization

We explore generalizations of some integrated learning and optimization frameworks for datadriven contextual stochastic optimization that can adapt to heteroscedasticity. We identify conditions on the stochastic program, data generation process, and the prediction setup under which these generalizations possess asymptotic and finite sample guarantees for a class of stochastic programs, including two-stage stochastic mixed-integer programs with continuous recourse. We verify that our assumptions hold for popular parametric and nonparametric regression methods.

[1]  Dimitris Bertsimas,et al.  From Predictive to Prescriptive Analytics , 2014, Manag. Sci..

[2]  Florian Ziel,et al.  Iteratively reweighted adaptive lasso for conditional heteroscedastic time series with applications to AR-ARCH type processes , 2015, Comput. Stat. Data Anal..

[3]  Christophe Chesneau,et al.  Nonparametric estimation in a regression model with additive and multiplicative noise , 2020, J. Comput. Appl. Math..

[4]  Anton Schick,et al.  Weighted least squares estimates in partly linear regression models , 1996 .

[5]  ADAPTIVE NONPARAMETRIC REGRESSION WITH CONDITIONAL HETEROSKEDASTICITY , 2014, Econometric Theory.

[6]  S. Sen,et al.  Learning Enabled Optimization : Towards a Fusion of Statistical Learning and Stochastic Programming , 2018 .

[7]  Sham M. Kakade,et al.  Random Design Analysis of Ridge Regression , 2012, COLT.

[8]  Priya L. Donti,et al.  Task-based End-to-end Model Learning in Stochastic Optimization , 2017, NIPS.

[9]  Heinrich Jiang,et al.  Non-Asymptotic Uniform Rates of Consistency for k-NN Regression , 2017, AAAI.

[10]  B. Hansen.  UNIFORM CONVERGENCE RATES FOR KERNEL ESTIMATION WITH DEPENDENT DATA , 2008, Econometric Theory.

[11]  A. Belloni,et al.  SPARSE MODELS AND METHODS FOR OPTIMAL INSTRUMENTS WITH AN APPLICATION TO EMINENT DOMAIN , 2012 .

[12]  Large and moderate deviations principles for kernel estimators of the multivariate regression , 2007, math/0703341.

[13]  David Ruppert,et al.  Robust Estimation in Heteroscedastic Linear Models. , 1982 .

[14]  Mohamed Hebiri,et al.  Learning Heteroscedastic Models by Convex Programming under Group Sparsity , 2013, ICML.

[15]  A. Belloni,et al.  Pivotal estimation via square-root Lasso in nonparametric regression , 2011, 1105.1475.

[16]  Jérémie Gallien,et al.  Dynamic Procurement of New Products with Covariate Information: The Residual Tree Method , 2019, Manuf. Serv. Oper. Manag..

[17]  James R. Luedtke,et al.  Residuals-based distributionally robust optimization with covariate information. , 2020 .

[18]  L. Bauwens,et al.  Multivariate GARCH Models: A Survey , 2003 .

[19]  James R. Luedtke,et al.  Data-Driven Sample Average Approximation with Covariate Information , 2022, 2207.13554.

[20]  Qiang Sun,et al.  Adaptive Huber Regression , 2017, Journal of the American Statistical Association.

[21]  N. Wermuth,et al.  Nonlinear Time Series : Nonparametric and Parametric Methods , 2005 .

[22]  R. Carroll,et al.  Variance Function Estimation , 1987 .

[23]  Eduardo F. Mendes,et al.  ℓ1-regularization of high-dimensional time-series models with non-Gaussian and heteroskedastic errors , 2016 .

[24]  David Ruppert,et al.  Local polynomial variance-function estimation , 1997 .

[25]  Jianqing Fan,et al.  A NEW PERSPECTIVE ON ROBUST M-ESTIMATION: FINITE SAMPLE THEORY AND APPLICATIONS TO DEPENDENCE-ADJUSTED MULTIPLE TESTING. , 2017, Annals of statistics.

[26]  Jianqing Fan,et al.  Quasi-Maximum Likelihood Estimation of GARCH Models With Heavy-Tailed Likelihoods , 2014 .

[27]  I. Keilegom,et al.  Semiparametric modeling and estimation of heteroscedasticity in regression analysis of cross-sectional data , 2010 .

[28]  P. Robinson.  Asymptotically efficient estimation in the presence of heteroskedasticity of unknown form , 1987 .

[29]  Alexander Shapiro,et al.  Lectures on Stochastic Programming: Modeling and Theory , 2009 .

[30]  Joseph P. Romano,et al.  Resurrecting Weighted Least Squares , 2016 .