IEEE Transactions on Information Theory · 1996 · 26 citations · 17 references
EngineeringMathematicsProbability TheoryStochastic ModelingStochastic ProcessesContinuous-time Stochastic ProcessesStochastic CalculusLinear ApproximationStochastic ApproximationWeak Convergence Of Stochastic Processes
Many stochastic approximation procedures result in a stochastic algorithm of the form h/sub k+1/=h/sub k/+1/k(b/sub k/-A/sub k/h/sub k/), for all k=1,2,3,... . Here, {b/sub k/,k=1,2,3...} is a R/sup d/-valued process, {A/sub k/,k=1,2,3,...} is a symmetric, positive semidefinite Re/sup d/spl times/d/-valued process, and {h/sub k/,k=1,2,3,...} is a sequence of stochastic estimates which hopefully converges to h/sup /spl Delta//=[lim/sub N/spl rarr//spl infin//1/N/spl Sigma//sub k=1//sup N/EA/sub k/]/sup -1/ {lim/sub N/spl rarr//spl infin//1/N/spl Sigma//sub k=1//sup N/Eb/sub k/} (assuming everything here is well defined). We give an elementary proof which relates the almost sure convergence of {h/sub k/,k=1,2,3,...} to strong laws of large numbers for {b/sub k/,k=1,2,3,...} and {A/sub k/,k=1,2,3,...}.
17
The algebraic eigenvalue problem
V.V. Voyevodin · USSR Computational Mathematics and Mathematical Physics · 1966 · 2.1K citations
M. H. D., William Stout · Journal of the American Statistical Association · 1975 · 1.3K citations
Asymptotics for Linear Processes
Peter C.B. Phillips, Victor Solo · The Annals of Statistics · 1992 · 913 citations · Full text
A Maximal Inequality and Dependent Strong Laws
D. L. McLeish · The Annals of Probability · 1975 · 435 citations · Full text