EconStor >
ifo Institut – Leibniz-Institut für Wirtschaftsforschung an der Universität München >
CESifo Working Papers, CESifo Group Munich >

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10419/19040
  
Title:Generalized stochastic gradient learning PDF Logo
Authors:Evans, George W.
Honkapohja, Seppo
Williams, Noah
Issue Date:2005
Series/Report no.:CESifo working papers 1576
Abstract:We study the properties of generalized stochastic gradient (GSG) learning in forward-looking models. We examine how the conditions for stability of standard stochastic gradient (SG) learning both differ from and are related to E-stability, which governs stability under least squares learning. SG algorithms are sensitive to units of measurement and we show that there is a transformation of variables for which E-stability governs SG stability. GSG algorithms with constant gain have a deeper justification in terms of parameter drift, robustness and risk sensitivity.
Subjects:adaptive learning
E-stability
recursive least squares
robust estimation
JEL:C65
C62
E17
E10
D83
Document Type:Working Paper
Appears in Collections:CESifo Working Papers, CESifo Group Munich

Files in This Item:
File Description SizeFormat
cesifo1_wp1576.pdf448.63 kBAdobe PDF
No. of Downloads: Counter Stats
Download bibliographical data as: BibTeX
Share on:http://hdl.handle.net/10419/19040

Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.