Premium
Parameter changes at unknown times in non‐linear regression
Author(s) -
Jandhyala Venkata K.,
AlSaleh Jamal A.
Publication year - 1999
Publication title -
environmetrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.68
H-Index - 58
eISSN - 1099-095X
pISSN - 1180-4009
DOI - 10.1002/(sici)1099-095x(199911/12)10:6<711::aid-env398>3.0.co;2-t
Subject(s) - mathematics , statistics , linear regression , proper linear model , linear model , bayesian multivariate linear regression , regression analysis , generalized linear model , general linear model , bayes' theorem , regression , econometrics , bayesian probability
Introducing the change‐point problem for the non‐linear regression model, we derive Bayes‐type statistics for tests of parameter changes at unknown times in the non‐linear parameters. We then establish asymptotic distribution theory, mainly, for the two‐sided statistics. Throughout the paper, we emphasize the specific case of an exponential type non‐linear regression model. The methodology is illustrated through data on pre‐school boys' weight/height ratio. Copyright © 1999 John Wiley & Sons, Ltd.