TWO STEPWISE REGRESSION METHODS AND CONSISTENT MODEL SELECTION FOR HIGHLY CORRELATED AND HIGH-DIMENSIONAL SPARSE LINEAR MODELS

Thesis

Fu, Yuan-Yi

Masters, 國立清華大學, 統計學研究所

2017

Abstract

弱正交貪婪演算法

多步正交貪婪演算法

高維度訊息準則

模型選擇

Weak orthogonal greedy algorithm

Multi-step orthogonal greedy algorithm

High-dimensional information criterion

Model selection

abstract hide

1 Record Views

Title: TWO STEPWISE REGRESSION METHODS AND CONSISTENT MODEL SELECTION FOR HIGHLY CORRELATED AND HIGH-DIMENSIONAL SPARSE LINEAR MODELS
Translated title: TWO STEPWISE REGRESSION METHODS AND CONSISTENT MODEL SELECTION FOR HIGHLY CORRELATED AND HIGH-DIMENSIONAL SPARSE LINEAR MODELS
Creators: Fu, Yuan-Yi
Contributors: Ing, Ching-Kang (Advisor)
Awarding Institution: 國立清華大學, 統計學研究所; Masters
Theses and Dissertations: Masters, 國立清華大學, 統計學研究所
Language: English
Resource Type: Thesis
Date published: 2018