PCR, PLS, and Lasso Regression

Andreas Tilevik
Additional contact information
Andreas Tilevik: University of Skövde

Chapter Chapter 13 in Multivariate Statistics and Machine Learning in R For Beginners, 2025, pp 279-297 from Springer

Abstract: Abstract This chapter begins by introducing the problem of multicollinearity, which occurs when predictor variables are highly correlated with each other. Multicollinearity can lead to unreliable coefficient estimates and inflated standard errors in methods, such as multiple linear regression. This issue is particularly problematic in high-dimensional datasets, where some predictors are likely to be strongly correlated. To address these challenges, we will explore how principal component regression and partial least squares regression combine highly correlated predictors, as well as how to compute confidence intervals using bootstrapping. This chapter ends with Lasso regression, a regularization technique that can shrink parameter estimates to zero. When multiple features are highly correlated, Lasso may randomly select one and discard the others.

Date: 2025
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-032-01851-9_13

Ordering information: This item can be ordered from
http://www.springer.com/9783032018519

DOI: 10.1007/978-3-032-01851-9_13

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().