Premium
A NOTE ON TRANSITION PROBABILITY ESTIMATION
Author(s) -
Dent Warren
Publication year - 1972
Publication title -
australian journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.434
H-Index - 41
eISSN - 1467-842X
pISSN - 0004-9581
DOI - 10.1111/j.1467-842x.1972.tb00896.x
Subject(s) - covariance , mathematics , estimation , variance (accounting) , statistics , quadratic equation , transition (genetics) , econometrics , economics , geometry , management , accounting , biochemistry , chemistry , gene
The usual regression model for the estimation of Markovian transition probabilities is examined under the assumption that the disturbance terms are normally distributed. It is found that for a certain variance‐covariance structure, the regular quadratic programming estimates are the same as the maximum likelihood estimates when non‐negativity is imposed.