Title :
Analytic first and second derivatives for the recursive prediction error algorithm´s log likelihood function
Author_Institution :
Dept. of Econ., Dartmouth Coll., Hanover, NH, USA
fDate :
3/1/1994 12:00:00 AM
Abstract :
This note improves on a method for computing analytic derivatives of the likelihood function of a discrete, linear, time invariant Gaussian state-space system and extends it to handle correlation between the state transition and measurement noise terms and to compute the analytic Hessian matrix
Keywords :
differentiation; discrete systems; error analysis; estimation theory; filtering and prediction theory; measurement errors; probability; state-space methods; analytic Hessian matrix; analytic first derivatives; analytic second derivatives; discrete linear time-invariant Gaussian state-space system; log likelihood function; measurement noise; recursive prediction error algorithm; state transition; Algorithm design and analysis; Control systems; Debugging; Error correction; Gaussian noise; Noise measurement; Prediction algorithms; Predictive models; Recursive estimation; Time measurement;
Journal_Title :
Automatic Control, IEEE Transactions on
