A FAMILY OF REGRESSION MODELS HAVING PARTIALLY ADDITIVE AND MULTIPLICATIVE COVARIATE STRUCTURE
スポンサーリンク
概要
- 論文の詳細を見る
We propose a new parametric regression model, Hybrid Linear Regression Model (or simply HLRM), which has partially additive and multiplicative covariate structure. In an ordinary linear regression model or generalized linear model, it is assumed that the covariates have either an additive or multiplicative effect on the response. A family of HLRM includes an ordinary linear regression model, logarithmic linear model and generalized linear model with normal errors as special cases. In analysis of HLRM, estimating unknown parameters or searching for the best fitting optimal model, we assume the log-normal distribution. Some illustrative analyses applying HLRM to actual data sets are also demonstrated.
- Research Association of Statistical Sciencesの論文
- 2005-12-00
Research Association of Statistical Sciences | 論文
- CLUSTERING BY A FUZZY METRIC : APPLICATIONS TO THE CLUSTER-MEDIAN PROBLEM
- A FAMILY OF REGRESSION MODELS HAVING PARTIALLY ADDITIVE AND MULTIPLICATIVE COVARIATE STRUCTURE
- AN OPTIMAL STOPPING PROBLEM ON TREE
- ON THE ORDERS OF MAX-MIN FUNCTIONALS
- TREE EXPRESSIONS AND THEIR PRODUCT FORMULA