PUBLICATIONS DE L'INSTITUT MATHÉMATIQUE (BEOGRAD) (N.S.) Vol. 70(84), pp. 6368 (2001) 

ESTIMATING OF PARAMETERS: NUAR(1) PROCESSMiroslav M. Risti\'c and Biljana \v C. Popovi\'cPrirodnomatemati\v cki fakultet, Ni\v s, YugoslaviaAbstract: We applied the method of conditional least squares for estimating parameters of NUAR(1). This process can be represented as the random coefficient autoregressive time series of the form $$ X_n=U_n X_{n1}+V_n, $$ where $\{(U_n,V_n)\}$ is the sequence of independent identically distributed random vectors such that supply the elements of the sequence $\{X_n\}$ with $\mathcal{U}(0,1)$ marginal distribution. Defined estimates were the functions of the estimates of moments $E(U_n)$ and $E(U_n V_n)$ and they are strong consistent and asymptotically normally distributed. Keywords: conditional least squares estimation; uniform autoregressive process; asymptotic normality; strong consistency Classification (MSC2000): 62M10 Full text of the article:
Electronic fulltext finalized on: 17 Oct 2002. This page was last modified: 13 Nov 2002.
© 2002 Mathematical Institute of the Serbian Academy of Science and Arts
