Skip to main navigation
Skip to search
Skip to main content
Princeton University Home
Help & FAQ
Home
Profiles
Research units
Facilities
Projects
Research output
Search by expertise, name or affiliation
CANONICAL THRESHOLDING FOR NONSPARSE HIGH-DIMENSIONAL LINEAR REGRESSION
Igor Silin,
Jianqing Fan
Operations Research & Financial Engineering
Bendheim Center for Finance
Center for Statistics & Machine Learning
Economics
Research output
:
Contribution to journal
›
Article
›
peer-review
Overview
Fingerprint
Fingerprint
Dive into the research topics of 'CANONICAL THRESHOLDING FOR NONSPARSE HIGH-DIMENSIONAL LINEAR REGRESSION'. Together they form a unique fingerprint.
Sort by
Weight
Alphabetically
Mathematics
Thresholding
100%
Linear regression
81%
High-dimensional
67%
Estimator
49%
Relative Error
36%
Decay
27%
Effective Dimension
23%
Principal Component Regression
22%
Fixed Design
22%
Random Design
22%
Eigenvalue
20%
Prediction Error
19%
Sparsity
17%
Canonical form
16%
Mean Squared Error
16%
Minimax
15%
Theoretical Analysis
15%
Family
15%
Covariance matrix
14%
Optimality
13%
Numerical Simulation
12%
Lower bound
10%
Performance
9%
Concepts
8%
Sufficient Conditions
8%
Form
6%
Business & Economics
Linear Regression
85%
Estimator
65%
Regression Coefficient
59%
Eigenvalues
41%
Decay
36%
Optimality
20%
Prediction Error
20%
Minimax
20%
Mean Squared Error
20%
Principal Components
18%
Covariance Matrix
18%
Numerical Simulation
17%
Lower Bounds
16%
Theoretical Analysis
16%
Performance
6%