Estimation of Nonorthogonal Problem Using Time Series Dataset

Abstract

Based on presentation of the principles of nonorthogonal problem, we discuss the difference of some approaches. A simple procedure to include the R-squared and Root Mean Square Error (R.M.S.E) is proposed and tested. The results showed that the Partial Least Square Regression provides better predictions due to a small R.M.S.E value.

Keyword. Nonorthogonal, Mean Square, Partial Least Square, R Square.

Table of Content

1.0 Introduction

2.0 The Ordinary Least Square Model

2.3 Ridge Regression (RR).

3.0 Result and Discussion

3.1 Illustrative Example

4.0 Conclusion

4.1 Recommendation

Reference