Search

Parameter estimation and the variable selection are two pioneer issues in regression analysis. While traditional variable selection methods require prior estimation of the model parameters, the penalized methods simultaneously carry on parameter estimation and variable select. Therefore, penalized variable selection methods are of great interest and have been extensively studied in literature. However, most of the papers in literature are only limited to the regression models with uncorrelated error terms and normality assumption. In this study, we combine the parameter estimation and the variable selection in regression models with autoregressive error term by using different penalty functions under heavy tailed error distribution assumption. We conduct a simulation study and a real data example to show the performance of the estimators.

In countries with a severe outbreak of COVID-19, most governments are considering whether anti-transmission measures are worth social and economic costs. The seriousness of economic costs such as the closure of some workplaces, unemployment, reduction in production, and social costs such as school closures, disruptions in education could be observable. However, the effect of the measures taken on the spread of the epidemic, such as the number of delayed or prevented cases, could not be observed. For this reason, the direct effects of the measures taken on health, that is, the effects on the course of the epidemic, are important research subjects. For this purpose, in this study, the breakpoint linear regression analysis is performed to analyze the trends of daily active cases, recovered, and deaths in Turkey. The analysis reveals that there has been a remarkable impact on lockdown and other precautions. Using the breakpoint regression model, we also analyze the active cases' trajectory for eight affected countries and compare the patterns in these countries with Turkey., Comment: 23 pages, 12 figures, 4 tables

Linear regression models are useful statistical tools to analyze data sets in several different fields. There are several methods to estimate the parameters of a linear regression model. These methods usually perform under normally distributed and uncorrelated errors with zero mean and constant variance. However, for some data sets error terms may not satisfy these or some of these assumptions. If error terms are correlated, such as the regression models with autoregressive (AR(p)) error terms, the Conditional Maximum Likelihood (CML) under normality assumption or the Least Square (LS) methods are often used to estimate the parameters of interest. For CML estimation a distributional assumption on error terms is needed to carry on estimation, but, in practice, such distributional assumptions on error terms may not be plausible. Therefore, in such cases some alternative distribution free methods are needed to conduct the parameter estimation. In this paper, we propose to estimate the parameters of a linear regression model with AR(p) error term using the Empirical Likelihood (EL) method, which is one of the distribution free estimation methods. A small simulation study and a numerical example are provided to evaluate the performance of the proposed estimation method over the CML method. The results of simulation study show that the proposed estimators based on EL method are remarkably better than the estimators obtained from the CML method in terms of mean squared errors (MSE) and bias in almost all the simulation configurations. These findings are also confirmed by the results of the numerical and real data examples., Comment: 18 pages, 3 figures, 5 tables

In the linear regression model with possibly autoregressive errors, we propose a family of nonparametric tests for regression under a nuisance autoregression. The tests avoid the estimation of nuisance parameters, in contrast to the tests proposed in the literature., Comment: 8 pages

Marshall-Olkin Extended Burr XII (MOEBXII) distribution family, which is a generalization of Burr XII distribution proposed by Al-Saiari et al. [1] , is a flexible distribution that can be used in many fields such as actuarial science, economics, life testing, reliability and failure time modeling. The parameters of the MOEBXII distribution are usually estimated by the maximum likelihood (ML) and least squares (LS) estimation methods. However, these estimators are not robust to the outliers which are often encountered in practice. There are two main purposes of this paper. The first one is to find the robust estimators for the parameters of the MOEBXII distribution. The second one is to use this distribution for modeling data from pharmacokinetics study. To obtain the robust estimators we use the optimal B robust estimator proposed by Hampel et al. [2]. We provide a simulation study to show the performance of the proposed estimators for the ML, LS and robust M estimators. We also give a real data example to illustrate the modeling capacity of the MOEBXII distribution for data from pharmacokinetics study., Comment: 27 pages, 4 figures

In this article, we consider the parameter estimation of regression model with pth order autoregressive (AR(p)) error term. We use the Maximum Lq-likelihood (MLq) estimation method that is proposed by Ferrari and Yang (2010a), as a robust alternative to the classical maximum likelihood (ML) estimation method to handle the outliers in the data. After exploring the MLq estimators for the parameters of interest, we provide some asymptotic properties of the resulting MLq estimators. We give a simulation study and a real data example to illustrate the performance of the new estimators over the ML estimators and observe that the MLq estimators have superiority over the ML estimators when outliers are present in the data.

In this article, we propose joint location, scale and skewness models of the skew Laplace normal (SLN) distribution as an alternative model for joint modelling location, scale and skewness models of the skew-t-normal (STN) distribution when the data set contains both asymmetric and heavy-tailed observations. We obtain the maximum likelihood (ML) estimators for the parameters of the joint location, scale and skewness models of the SLN distribution using the expectation-maximization (EM) algorithm. The performance of the proposed model is demonstrated by a simulation study and a real data example.

Ordinary least square (OLS), maximum likelihood (ML) and robust methods are the widely used methods to estimate the parameters of a linear regression model. It is well known that these methods perform well under some distributional assumptions on error terms. However, these distributional assumptions on the errors may not be appropriate for some data sets. In these case, nonparametric methods may be considered to carry on the regression analysis. Empirical likelihood (EL) method is one of these nonparametric methods. The EL method maximizes a function, which is multiplication of the unknown probabilities corresponding to each observation, under some constraints inherited from the normal equations in OLS estimation method. However, it is well known that the OLS method has poor performance when there are some outliers in the data. In this paper, we consider the EL method with robustifyed constraints. The robustification of the constraints is done by using the robust M estimation methods for regression. We provide a small simulation study and a real data example to demonstrate the capability of the robust EL method to handle unusual observations in the data. The simulation and real data results reveal that robust constraints are needed when heavy tailedness and/or outliers are possible in the data., Comment: 15 pages, 6 tables, 1 figure

In this paper, we consider a linear regression model with AR(p) error terms with the assumption that the error terms have a t distribution as a heavy tailed alternative to the normal distribution. We obtain the estimators for the model parameters by using the conditional maximum likelihood (CML) method. We conduct an iteratively reweighting algorithm (IRA) to find the estimates for the parameters of interest. We provide a simulation study and three real data examples to illustrate the performance of the proposed robust estimators based on t distribution.

The use of prior information in the linear regression is well known to provide more efficient estimators of regression coefficients. The methods of non-stochastic restricted regression estimation proposed by Theil and Goldberger (1961) are preferred when prior information is available. In this study, we will consider parameter estimation and the variable selection in non-stochastic restricted linear regression model, using least absolute shrinkage and selection operator (LASSO) method introduced by Tibshirani (1996). A small simulation study and real data example are provided to illustrate the performance of the proposed method for dealing with the variable selection and the parameter estimation in restricted linear regression models., Comment: 12 pages, 4 figures 6 tables

The normality assumption on data set is very restrictive approach for modelling. The generalized form of normal distribution, named as an exponential power (EP) distribution, and its scale mixture form have been considered extensively to overcome the problem for modelling non-normal data set since last decades. However, examining the robustness properties of maximum likelihood (ML) estimators of parameters in these distributions, such as the in uence function, gross-error sensitivity, breakdown point and information-standardized sensitivity, has not been considered together. The well-known asymptotic properties of ML estimators of location, scale and added skewness parameters in EP and its scale mixture form distributions are studied and also these ML estimators for location, scale and scale variant (skewness) parameters can be represented as an iterative reweighting algorithm to compute the estimates of these parameters simultaneously., Comment: 28 pages 4 figures

The t-distribution has many useful applications in robust statistical analysis. The parameter estimation of the t-distribution is carried out using ML estimation method, and the ML estimates are obtained via the EM algorithm. In this study, we consider an alternative estimation method for all the parameters of the multivariate-t distribution using the MLq estimation method. We adapt the EM algorithm to obtain the MLq estimates for all the parameters. We provide a small simulation study to illustrate the performance of the MLq estimators over the ML estimators and observe that the MLq estimators have considerable superiority over the ML estimators., Comment: 14 pages, 4 figures

The Burr III distribution is used in a wide variety of fields of lifetime data analysis, reliability theory, and financial literature, etc. It is defined on the positive axis and has two shape parameters, say $c$ and $k$. These shape parameters make the distribution quite flexible. They also control the tail behavior of the distribution. In this study, we extent the Burr III distribution to the real axis and also add a skewness parameter, say $\varepsilon$, with epsilon-skew extension approach. When the parameters $c$ and $k$ have a relation such that $ck \approx 1 $ or $ck < 1 $, it is skewed unimodal. Otherwise, it is skewed bimodal with the same level of peaks on the negative and positive sides of real line. Thus, ESBIII distribution can capture fitting the various data sets even when the number of parameters are three. Location and scale form of this distribution are also given. Some distributional properties of the new distribution are investigated. The maximum likelihood (ML) estimation method for the parameters of ESBIII is considered. The robustness properties of ML estimators are studied and also tail behaviour of ESBIII distribution is examined. The applications on real data are considered to illustrate the modeling capacity of this distribution in the class of bimodal distributions., Comment: 22 pages, 5 figures

In this paper, we propose finite mixtures of multivariate skew Laplace distributions to model both skewness and heavy-tailedness in the heterogeneous data sets. The maximum likelihood estimators for the parameters of interest are obtained by using the EM algorithm. We give a small simulation study and a real data example to illustrate the performance of the proposed mixture model.

The asymmetric objective function is proposed as an alternative to Huber objective function to model skewness and obtain robust estimators for the location, scale and skewness parameters. The robustness and asymptotic properties of the asymmetric M-estimators are explored. A simulation study and real data examples are given to illustrate the performance of proposed asymmetric M-estimation method over the symmetric M-estimation method. It is observed from the simulation results that the asymmetric M-estimators perform better than Huber M-estimators when the data have skewness. The application on regression is also considered., Comment: 31 pages

In this study, we propose a robust mixture regression procedure based on the skew t distribution to model heavy-tailed and/or skewed errors in a mixture regression setting. Using the scale mixture representation of the skew t distribution, we give an Expectation Maximization (EM) algorithm to compute the maximum likelihood (ML) estimates for the paramaters of interest. The performance of proposed estimators is demonstrated by a simulation study and a real data example., Comment: 15

The bridge regression estimator generalizes both ridge regression and LASSO estimators. Since it minimizes the sum of squared residuals with a $L_{\gamma }$ penalty, this estimator is typically not robust against outliers in the data. There have been attempts to define robust versions of the bridge regression method, but while these proposed methods produce bridge regression estimators robust to outliers and heavy-tailed errors, they are not robust against leverage points. We propose a robust bridge regression estimation method combining MM and bridge regression estimation methods. The MM bridge regression estimator obtained from the proposed method is robust against outliers and leverage points. Furthermore, for appropriate choices of the penalty function, the proposed method is able to perform variable selection and parameter estimation simultaneously. Consistency, asymptotic normality, and sparsity of the MM bridge regression estimator are achieved. We propose an algorithm to compute the MM bridge regression estimate. A simulation study and a real data example are provided to demonstrate the performance of the MM bridge regression estimator for finite sample cases.

Bai (2010) and Bai et al. (2012) proposed robust mixture regression method based on the M regression estimation. However, the M-estimators are robust against the outliers in response variables, but they are not robust against the outliers in explanatory variables (leverage points). In this paper, we propose a robust mixture regression procedure to handle the outliers and the leverage points, simultaneously. Our proposed mixture regression method is based on the GM regression estimation. We give an Expectation Maximization (EM) type algorithm to compute estimates for the parameters of interest. We provide a simulation study and a real data example to assess the robustness performance of the proposed method against the outliers and the leverage points., Comment: 15 pages