Latent Factor Prediction Pursuit for Rank Deficient Regressors
نویسنده
چکیده
In simulation studies Latent Factor Prediction Pursuit outperformed classical reduced rank regression methods. The algorithm described so far for Latent Factor Prediction Pursuit had two shortcomings: It was only implemented for situations where the explanatory variables were of full colum rank. Also instead of the projection matrix only the regression matrix was calculated. These problems are addressed by a new algorithm which finds the prediction optimal projection.
منابع مشابه
Simple Estimators for Binary Choice Models With Endogenous Regressors
This paper provides two main contributions to binary choice models with endogenous regressors. First, we propose some variants of special regressor based estimators that are numerically trivial to implement. These estimators provide consistent estimates of binary choice model coef cients when regressors (either discretely or continuously distributed) are endogenous, and when the latent errors h...
متن کاملBayesian Probabilistic Multi-Topic Matrix Factorization for Rating Prediction
Recently, Local Matrix Factorization (LMF) [Lee et al., 2013] has been shown to be more effective than traditional matrix factorization for rating prediction. The core idea for LMF is to first partition the original matrix into several smaller submatrices, further exploit local structures of submatrices for better low-rank approximation. Various clustering-based methods with heuristic extension...
متن کاملAn Improved Independent Component Regression Modeling and Quantitative Calibration Procedure
An improved independent component regression (M-ICR) algorithm is proposed by constructing joint latent variable (LV) based regressors, and a quantitative statistical analysis procedure is designed using a bootstrap technique for model validation and performance evaluation. First, the drawbacks of the conventional regression modeling algorithms are analyzed. Then the proposed M-ICR algorithm is...
متن کاملOrthogonal Rank-One Matrix Pursuit for Low Rank Matrix Completion
In this paper, we propose an efficient and scalable low rank matrix completion algorithm. The key idea is to extend orthogonal matching pursuit method from the vector case to the matrix case. We further propose an economic version of our algorithm by introducing a novel weight updating rule to reduce the time and storage complexity. Both versions are computationally inexpensive for each matrix ...
متن کاملPosterior Means and Precisions of the Coeffi cients in Linear Models with Highly Collinear Regressors∗
When there is exact collinearity between regressors, their individual coeffi cients are not identified, but given an informative prior their Bayesian posterior means are well defined. The case of high but not exact collinearity is more complicated but similar results follow. Just as exact collinearity causes non-identification of the parameters, high collinearity can be viewed as weak identific...
متن کامل