New Methods of Enhancing Prediction Accuracy in Linear Models with Missing Data
This addresses prediction accuracy issues for linear models with missing data, but appears incremental as it modifies existing algorithms for missing scenarios.
The paper tackles prediction in linear models with missing data by introducing new methods to improve Mean Squared Error (MSE) on test sets compared to state-of-the-art approaches, focusing on tuning the Bias-Variance trade-off; simulation results confirm these heuristics enhance prediction accuracy.
In this paper, prediction for linear systems with missing information is investigated. New methods are introduced to improve the Mean Squared Error (MSE) on the test set in comparison to state-of-the-art methods, through appropriate tuning of Bias-Variance trade-off. First, the use of proposed Soft Weighted Prediction (SWP) algorithm and its efficacy are depicted and compared to previous works for non-missing scenarios. The algorithm is then modified and optimized for missing scenarios. It is shown that controlled over-fitting by suggested algorithms will improve prediction accuracy in various cases. Simulation results approve our heuristics in enhancing the prediction accuracy.