Premium
On indirect least squares estimation of a simultaneous equation system
Author(s) -
Park SooBin
Publication year - 1974
Publication title -
canadian journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.804
H-Index - 51
eISSN - 1708-945X
pISSN - 0319-5724
DOI - 10.2307/3314964
Subject(s) - estimator , least squares function approximation , mathematics , total least squares , generalized least squares , simultaneous equations model , structural equation modeling , least trimmed squares , linear least squares , non linear least squares , partial least squares regression , explained sum of squares , statistics , regression analysis
The indirect least squares ( ILS ) approach is broadly regarded as a procedure of deriving the structural parameters of a simultaneous equation system indirectly from the estimates of reduced‐form parameters. In this paper it is shown that the ILS approach yields estimators which are identical with the two‐stage least squares estimator, when applied to a single equation, and the three‐stage least squares estimator, when applied to the complete system. Thus a fresh motivation for the ILS method is provided.