In this article we propose a general Bayesian estimation framework for the Birnbaum-Saunders Quantile Regression (BSQ) model which is used to analyze asymmetric and heterogeneous economic data. In this work, the novel aspect is the joint incorporation of several existing surrogate structures (with different degrees of approximation) and decision-theoretic loss functions (with direct impact on estimation performance trade-offs) under a common BSQ framework conducive to systematic testing of their effect on quantile estimation performance. We consider two previous prior specifications which are Jeffreys prior and weakly informative priors together with three loss functions quadratic, LINEX and the proposed quadratic-logarithmic loss function. Iraqi economic data analysis demonstrates significant disparities in independent variable impacts across quantiles, confirming heterogeneous distribution. The results show further that Jeffreys prior yields more consistent estimates in most quantiles, whereas weakly informative priors give better coverage under the tail. Also, this highlights the importance of choosing a loss function and motivates asymmetric and hybrid loss functions for improved accuracy, especially in extreme quantiles. This contribution expands upon previous works by introducing a new general comparative framework that overextends the potential of Bayesian BSQ modelling, providing robustness and flexibility in such context with significant practical function resulting estimation improves in economic real-world applications.
Spatial data observed on a group of areal units is common in scientific applications. The usual hierarchical approach for modeling this kind of dataset is to introduce a spatial random effect with an autoregressive prior. However, the usual Markov chain Monte Carlo scheme for this hierarchical framework requires the spatial effects to be sampled from their full conditional posteriors one-by-one resulting in poor mixing. More importantly, it makes the model computationally inefficient for datasets with large number of units. In this article, we propose a Bayesian approach that uses the spectral structure of the adjacency to construct a low-rank expansion for modeling spatial dependence. We propose a pair of computationally efficient estimati
... Show More
(10)
(7)
(16)
(18)
In general, researchers and statisticians in particular have been usually used non-parametric regression models when the parametric methods failed to fulfillment their aim to analyze the models precisely. In this case the parametic methods are useless so they turn to non-parametric methods for its easiness in programming. Non-parametric methods can also used to assume the parametric regression model for subsequent use. Moreover, as an advantage of using non-parametric methods is to solve the problem of Multi-Colinearity between explanatory variables combined with nonlinear data. This problem can be solved by using kernel ridge regression which depend o
... Show More
TWMS Journal of Applied and Engineering Mathematics | Volume: 16 Issue: 9
Most statistical research generally relies on the study of the behaviour of different phenomena during specific time periods and the use of the results of these studies in the development of appropriate recommendations and decision-making and for the purpose of statistical inference on the parameters of the statistical distribution of life times in The technical staff of most of the manufacturers in the research units of these companies deals with censored data, the main objective of the study of survival is the need to provide information that is the basis for decision making and must clarify the problem and then the goals and limitations of this study and that It may have different possibilities to perform the
... Show More
The aim of the research is to estimate the hidden population. Here، the number of drug users in Baghdad was calculated for the male age group (15-60) years old ، based on the Bayesian models. These models are used to treat some of the bias in the Killworth method Accredited in many countries of the world.
Four models were used: random degree، Barrier effects، Transmission bias، the first model being random، an extension of the Killworth model، adding random effects such as variance and uncertainty Through the size of the personal network، and when expanded by adding the fact that the respondents have different tendencies، the mixture of non-random variables with random to produce
... Show More
The use of deep learning.
Researchers need to understand the differences between parametric and nonparametric regression models and how they work with available information about the relationship between response and explanatory variables and the distribution of random errors. This paper proposes a new nonparametric regression function for the kernel and employs it with the Nadaraya-Watson kernel estimator method and the Gaussian kernel function. The proposed kernel function (AMS) is then compared to the Gaussian kernel and the traditional parametric method, the ordinary least squares method (OLS). The objective of this study is to examine the effectiveness of nonparametric regression and identify the best-performing model when employing the Nadaraya-Watson
... Show More
(8)
(3)
Information security in data storage and transmission is increasingly important. On the other hand, images are used in many procedures. Therefore, preventing unauthorized access to image data is crucial by encrypting images to protect sensitive data or privacy. The methods and algorithms for masking or encoding images vary from simple spatial-domain methods to frequency-domain methods, which are the most complex and reliable. In this paper, a new cryptographic system based on the random key generator hybridization methodology by taking advantage of the properties of Discrete Cosine Transform (DCT) to generate an indefinite set of random keys and taking advantage of the low-frequency region coefficients after the DCT stage to pass them to
... Show More
(9)
(2)
In many scientific fields, Bayesian models are commonly used in recent research. This research presents a new Bayesian model for estimating parameters and forecasting using the Gibbs sampler algorithm. Posterior distributions are generated using the inverse gamma distribution and the multivariate normal distribution as prior distributions. The new method was used to investigate and summaries Bayesian statistics' posterior distribution. The theory and derivation of the posterior distribution are explained in detail in this paper. The proposed approach is applied to three simulation datasets of 100, 300, and 500 sample sizes. Also, the procedure was extended to the real dataset called the rock intensity dataset. The actual dataset is collecte
... Show More
(2)