imprint:
Cambridge, Mass: National Bureau of Economic Research, October 2005
Published in:NBER technical working paper series ; no. t0317
Extent:
1 Online-Ressource
Language:
English
DOI:
10.3386/t0317
Identifier:
Reproduction note:
Hardcopy version available to institutional subscribers
Origination:
Footnote:
Mode of access: World Wide Web
System requirements: Adobe [Acrobat] Reader required for PDF files
Description:
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