TY - RPRT
T1 - Estimating Functions with Prior Knowledge, (EFPK) for diffusions
AU - Nolsøe, Kim
AU - Kessler, Mathieu
AU - Madsen, Henrik
PY - 2003
Y1 - 2003
N2 - In this paper a method is formulated in an estimating function setting for parameter estimation, which allows the use of prior information. The main idea is to use prior knowledge of the parameters, either specified as moments restrictions or as a distribution, and use it in the construction of an estimating function. It may be useful when the full Bayesian analysis is difficult to carry out for computational reasons. This is almost always the case for diffusions, which is the focus of this paper, though the method applies in other settings.
AB - In this paper a method is formulated in an estimating function setting for parameter estimation, which allows the use of prior information. The main idea is to use prior knowledge of the parameters, either specified as moments restrictions or as a distribution, and use it in the construction of an estimating function. It may be useful when the full Bayesian analysis is difficult to carry out for computational reasons. This is almost always the case for diffusions, which is the focus of this paper, though the method applies in other settings.
KW - Diffusion Process
KW - Small sample size
KW - Ornstein-Uhlenbeck Process
KW - Estimating Functions
KW - Cox Ingersoll & Ross (CIR) Process
M3 - Report
BT - Estimating Functions with Prior Knowledge, (EFPK) for diffusions
PB - Informatics and Mathematical Modelling, Technical University of Denmark, DTU
ER -