All the identification algorithms give asymptotically unbiased estimates with general inputs, and the rate of reduction of bias can be estimated. Simulation results also show that the new algorithms converge much more rapidly (with sample size) than the existing method, and hence are more effective with small sample sizes. These advantages are achieved by a different arrangement of the input-output equations into `blocks', and projections onto different spaces than the ones used in the existing method. A further advantage of our algorithms is that the dimensions of the matrices involved are significantly smaller, so that the computational complexity is lower.
Back to Control Seminars Page.