Bootstrapping Principal Component Regression Models
نویسنده
چکیده
Bootstrap methods can be used as an alternative for cross-validation in regression procedures such as principal component regression (PCR). Several bootstrap methods for the estimation of prediction errors and confidence intervals are presented. It is shown that bootstrap error estimates are consistent with cross-validation estimates but exhibit less variability. This makes it easier to select the correct number of latent variables in the model. Using bootstrap confidence intervals for the regression vectors, it is possible to select a subset of the original variables to include in the regression, yielding a more parsimonious model with smaller prediction errors. The methods are illustrated using PCR, but can be applied to all regression models yielding a vector or matrix of regression coefficients. © 1997 by John Wiley & Sons, Ltd.
منابع مشابه
Derivation of regression models for pan evaporation estimation
Evaporation is an essential component of hydrological cycle. Several meteorologicalfactors play role in the amount of pan evaporation. These factors are often related to eachother. In this study, a multiple linear regression (MLR) in conjunction with PrincipalComponent Analysis (PCA) was used for modeling of pan evaporation. After thestandardization of the variables, independent components were...
متن کاملFunctional Data Analysis Applied to Modeling of Severe Acute Mucositis and Dysphagia Resulting From Head and Neck Radiation Therapy
PURPOSE Current normal tissue complication probability modeling using logistic regression suffers from bias and high uncertainty in the presence of highly correlated radiation therapy (RT) dose data. This hinders robust estimates of dose-response associations and, hence, optimal normal tissue-sparing strategies from being elucidated. Using functional data analysis (FDA) to reduce the dimensiona...
متن کاملKriging uncertainty for functional data: a comparison study
Uncertainty evaluation for spatial prediction of curves remains an open issue in the functional data literature. We consider three different approaches that rely on semi-parametric bootstrapping, principal component analysis and classical inference for additive models respectively.
متن کاملPredicting the Young\'s Modulus and Uniaxial Compressive Strength of a typical limestone using the Principal Component Regression and Particle Swarm Optimization
In geotechnical engineering, rock mechanics and engineering geology, depending on the project design, uniaxial strength and static Youngchr('39')s modulus of rocks are of vital importance. The direct determination of the aforementioned parameters in the laboratory, however, requires intact and high-quality cores and preparation of their specimens have some limitations. Moreover, performing thes...
متن کاملFunctional time series forecasting
We propose forecasting functional time series using weighted functional principal component regression and weighted functional partial least squares regression. These approaches allow for smooth functions, assign higher weights to more recent data, and provide a modeling scheme that is easily adapted to allow for constraints and other information. We illustrate our approaches using age-specific...
متن کامل