EconBiz - Find Economic Literature
    • Logout
    • Change account settings
  • A-Z
  • Beta
  • About EconBiz
  • News
  • Thesaurus (STW)
  • Academic Skills
  • Help
  •  My account 
    • Logout
    • Change account settings
  • Login
EconBiz - Find Economic Literature
Publications Events
Search options
Advanced Search history
My EconBiz
Favorites Loans Reservations Fines
    You are here:
  • Home
  • Search: subject:"EM-type algorithms"
Narrow search

Narrow search

Year of publication
Subject
All
EM-type algorithms 2 Cox proportional hazards model 1 Eigenvalue decomposition 1 F–G algorithm 1 Integrated completed likelihood 1 Logistic regression 1 Model-based clustering 1 Multivariate t mixture models 1 Newton–Raphson algorithm 1 Optimal QLB algorithm 1 QLB algorithm 1
more ... less ...
Online availability
All
Undetermined 2
Type of publication
All
Article 2
Language
All
Undetermined 2
Author
All
Lin, Tsung-I 1 Liu, Chunling 1 Tang, Man-Lai 1 Tian, Guo-Liang 1
Published in...
All
Computational Statistics & Data Analysis 2
Source
All
RePEc 2
Showing 1 - 2 of 2
Cover Image
Learning from incomplete data via parameterized t mixture models through eigenvalue decomposition
Lin, Tsung-I - In: Computational Statistics & Data Analysis 71 (2014) C, pp. 183-195
EM-type algorithms are developed for parameter estimation of these models under a missing at random (MAR) mechanism. For …
Persistent link: https://www.econbiz.de/10010871421
Saved in:
Cover Image
Accelerating the quadratic lower-bound algorithm via optimizing the shrinkage parameter
Tian, Guo-Liang; Tang, Man-Lai; Liu, Chunling - In: Computational Statistics & Data Analysis 56 (2012) 2, pp. 255-265
When the Newton–Raphson algorithm or the Fisher scoring algorithm does not work and the EM-type algorithms are not … available, the quadratic lower-bound (QLB) algorithm may be a useful optimization tool. However, like all EM-type algorithms …
Persistent link: https://www.econbiz.de/10011056438
Saved in:
A service of the
zbw
  • Sitemap
  • Plain language
  • Accessibility
  • Contact us
  • Imprint
  • Privacy

Loading...