This paper proposes a new difference-based estimation method for estimating the semi parametric p... more This paper proposes a new difference-based estimation method for estimating the semi parametric partially linear model (PLM). This method is called the difference for difference (DFD) estimation method, which is proposed by the author, for estimating the residual variance in nonparametric regression models. In this work, the DFD estimation method is used for estimating both the parametric component and the residual variance. A numerical study has been shown that the proposed DFD estimation gives best results compared to other existing difference methods; in the form of less mean squared error of parametric component and less residual variance of the fitted model.
In this article, estimation methods of the semiparametric generalized linear model known as the g... more In this article, estimation methods of the semiparametric generalized linear model known as the generalized partial linear model (GPLM) are reviewed. These methods are based on using kernel smoothing functions in the estimation of the nonparametric component of the model. We derive the algorithms for the estimation process and develop these algorithms for the generalized partial linear model (GPLM) with a binary response.
In this article, estimation methods of the semiparametric generalized linear model known as the g... more In this article, estimation methods of the semiparametric generalized linear model known as the generalized partial linear model (GPLM) are reviewed. These methods are based on using kernel smoothing functions in the estimation of the nonparametric component of the model. We derive the algorithms for the estimation process and develop these algorithms for the generalized partial linear model (GPLM) with a binary response.
In this paper, a semiparametric generalized linear model known as a generalized partial linear mo... more In this paper, a semiparametric generalized linear model known as a generalized partial linear model (GPLM) is estimated using the profile likelihood method. The Algorithms of the estimation process is derived and applied to credit scoring data. Credit scoring data is a typical example for the application of GPLM with a binomially distributed response variable. Key Wards: credit scoring, generalized linear model, generalized partial linear model, misclassification rates, profile-likelihood, quasi-likelihood, semiparametric estimation.
More attention has been given to regularization methods in the last two decades as a result of ex... more More attention has been given to regularization methods in the last two decades as a result of exiting high-dimensional ill-posed data. This paper proposes a new method of introducing the penalized term in regularized regression. The proposed penalty is based on using the least squares estimator’s variances of the regression parameters. The proposed method is applied to some penalized estimators like ridge, lasso, and elastic net, which are used to overcome both the multicollinearity problem and selecting variables. Good results are obtained using the average mean squared error criterion (AMSE) for simulated data, also real data are shown best results in the form of less average prediction errors (APE) of the resulting estimators. Keywords: Elastic-Net, Lasso, Penalized regression; Regularization; Ridge regression; Shrinkage; Variable selection AMS 2010 Mathematics Subject Classification : 62J05; 62J07
In this paper, a semiparametric generalized linear model known as a generalized partial linear mo... more In this paper, a semiparametric generalized linear model known as a generalized partial linear model (GPLM) is estimated using the profile likelihood method. The Algorithms of the estimation process is derived and applied to credit scoring data. Credit scoring data is a typical example for the application of GPLM with a binomially distributed response variable.
In this article, estimation methods of the semiparametric generalized linear model known as the g... more In this article, estimation methods of the semiparametric generalized linear model known as the generalized partial linear model (GPLM) are reviewed. These methods are based on using kernel smoothing functions in the estimation of the nonparametric component of the model. We derive the algorithms for the estimation process and develop these algorithms for the generalized partial linear model (GPLM) with
In this paper, a proposed form of dependent regression models are introduced to study symbolic da... more In this paper, a proposed form of dependent regression models are introduced to study symbolic data. The estimation of the proposed linear regression models are based on interval valued data, for which we have lower and upper bounds or center and range values. The least squares method is used to estimate the models. A real example data are used to illustrate the usefulness of the proposed regression models for handling the interval valued data. The estimation results are evaluated using the predicted mean squared errors. The results support the proposed dependent regression models.
This paper proposes a new difference-based estimation method for estimating the semi parametric p... more This paper proposes a new difference-based estimation method for estimating the semi parametric partially linear model (PLM). This method is called the difference for difference (DFD) estimation method, which is proposed by the author, for estimating the residual variance in nonparametric regression models. In this work, the DFD estimation method is used for estimating both the parametric component and the residual variance. A numerical study has been shown that the proposed DFD estimation gives best results compared to other existing difference methods; in the form of less mean squared error of parametric component and less residual variance of the fitted model.
In this article, estimation methods of the semiparametric generalized linear model known as the g... more In this article, estimation methods of the semiparametric generalized linear model known as the generalized partial linear model (GPLM) are reviewed. These methods are based on using kernel smoothing functions in the estimation of the nonparametric component of the model. We derive the algorithms for the estimation process and develop these algorithms for the generalized partial linear model (GPLM) with a binary response.
In this article, estimation methods of the semiparametric generalized linear model known as the g... more In this article, estimation methods of the semiparametric generalized linear model known as the generalized partial linear model (GPLM) are reviewed. These methods are based on using kernel smoothing functions in the estimation of the nonparametric component of the model. We derive the algorithms for the estimation process and develop these algorithms for the generalized partial linear model (GPLM) with a binary response.
In this paper, a semiparametric generalized linear model known as a generalized partial linear mo... more In this paper, a semiparametric generalized linear model known as a generalized partial linear model (GPLM) is estimated using the profile likelihood method. The Algorithms of the estimation process is derived and applied to credit scoring data. Credit scoring data is a typical example for the application of GPLM with a binomially distributed response variable. Key Wards: credit scoring, generalized linear model, generalized partial linear model, misclassification rates, profile-likelihood, quasi-likelihood, semiparametric estimation.
More attention has been given to regularization methods in the last two decades as a result of ex... more More attention has been given to regularization methods in the last two decades as a result of exiting high-dimensional ill-posed data. This paper proposes a new method of introducing the penalized term in regularized regression. The proposed penalty is based on using the least squares estimator’s variances of the regression parameters. The proposed method is applied to some penalized estimators like ridge, lasso, and elastic net, which are used to overcome both the multicollinearity problem and selecting variables. Good results are obtained using the average mean squared error criterion (AMSE) for simulated data, also real data are shown best results in the form of less average prediction errors (APE) of the resulting estimators. Keywords: Elastic-Net, Lasso, Penalized regression; Regularization; Ridge regression; Shrinkage; Variable selection AMS 2010 Mathematics Subject Classification : 62J05; 62J07
In this paper, a semiparametric generalized linear model known as a generalized partial linear mo... more In this paper, a semiparametric generalized linear model known as a generalized partial linear model (GPLM) is estimated using the profile likelihood method. The Algorithms of the estimation process is derived and applied to credit scoring data. Credit scoring data is a typical example for the application of GPLM with a binomially distributed response variable.
In this article, estimation methods of the semiparametric generalized linear model known as the g... more In this article, estimation methods of the semiparametric generalized linear model known as the generalized partial linear model (GPLM) are reviewed. These methods are based on using kernel smoothing functions in the estimation of the nonparametric component of the model. We derive the algorithms for the estimation process and develop these algorithms for the generalized partial linear model (GPLM) with
In this paper, a proposed form of dependent regression models are introduced to study symbolic da... more In this paper, a proposed form of dependent regression models are introduced to study symbolic data. The estimation of the proposed linear regression models are based on interval valued data, for which we have lower and upper bounds or center and range values. The least squares method is used to estimate the models. A real example data are used to illustrate the usefulness of the proposed regression models for handling the interval valued data. The estimation results are evaluated using the predicted mean squared errors. The results support the proposed dependent regression models.
Uploads