IRAQI JOURNAL OF STATISTICAL SCIENCES

In this paper, two lomax distribution parameters are estimated along with the estimation of the reliability function under two balanced loss functions: the balanced squared error function and balanced linex loss function. These two functions depend on both Bayesian and maximum likelihood estimators using one type of generalized order statistics, which is the upper recorded values.
        The simulation approach using matlab language program is adopted in order to generate the data and compute the estimators.
        The comparison between shape parameter ( ) estimation methods is done by using posterior Bayesian risk function. The findings show that the estimators under two balanced loss functions are more efficient than the estimators under the two ordinary loss functions .

Reliability growth is defined as the positive progress in reliability over a period of time according to changes in product design or manufacturing processes. There are three main areas of reliability growth: planning, tracking and projecting. The most important model is the tracking models The model parameters were estimated using the maximum potential method as well as estimation of the confidence intervals for these parameters, estimation of the confidence intervals for the time-to-time ratio (DMTBF), and finally the model was simulated using the Monte Carlo method using specialized software (RGA)

This paper compares a prediction accuracy between the statistical time series method that use (Box – Jenkins) methodology, and some artificial neural networks, which differ from them through the feedback in their structure.
These networks are Feed Forward Neural Network (FFNN), Elman Neural Network (ENN), and Nonlinear Autoregressive with Exogenous Input (NARX). By using a set of data, the average monthly maximum temperatures in Mosul for the years (1983– 2009), which numbered 324 observations, twelve observations were kept as Truncated samples in order to compare with the results of prediction models for the above two methods. The results of prediction with performance of neural networks with feedback is better than others, and the performance methodology of (Box - Jenkins).

The aim of this research is to compare the bootstrap confidence intervals with the Bayesian confidence intervals for smoothing splines as well as the traditional confidence intervals to determine which of these limits are best in the presence of Outliers and Leverage points in data. The simulation experiments were conducted on two models: the first was linear in the presence of data that was contaminated with outliers and the other with the Leverage points: The second model was nonlinear in the presence of data contaminated with outlying observations 
Simulation experiments were carried out on different samples. The Penalized Least Squares method was used to fit the Nonparametric regression. The Generalized Cross Validation function (GCV) was used to select the amount of smoothing.

In many kinds of pollution, such as economic and environmental pollution, the researchers use the normal linear model to present  their data studies. That selection may be inaccurate because the data of  those studies do not vacate from outlier observations, which have great effect on the estimation problem even if they are processed or removed from the sample study. These processes lead to facts defacement to the decision maker. For that reason, the non-normal linear models has been found out to combat that matter. That error term in these models belongs to the family of probability distributions which resist outliers, for example, the multivariate t and mixture normal distributions. 
       The factor analysis model belongs to the family of linear models and because the multivariate data sets do not vacate outliers .For this reason this paper is concerned with studying the t factor analysis model. The model analyzed by Bayesian technique in which the common factors are treated as fixed and random variables . We supposed that all parameters of both two models were unknown and their prior distributions belong to conjugate families.
     The number of extracted factors in factor analysis models cannot be determined a prior .On this foundation, in Bayesian analysis, these factors are treated as random variables. We obtained a posterior probability criterion to choose the number of extracted factors for the two models. We choose the number of factors in which they must be entered, and the model which they have maximum posterior probability.
     All results that we concluded were applied to empirical data sets which are generated by simulation in two different sample sizes (n=50,100) at different values of the degrees of freedom for the distribution of the error term. Also, we selected different forms of factor loading matrix and variance matrix of error term. Matlab (7.9) language is used in data generation and analysis

The last few years witnessed a great and increasing interest in the field of intelligent classification techniques which rely on Machine Learning. In recent times Machine Learning one of the areas in Artificial Intelligence has been widely used in order to assist medical experts and doctors in the prediction and diagnosis of different diseases. In this paper, we applied two different machine learning algorithms to a problem in the domain of medical diagnosis and analyzed their efficiency in prediction the results. The problem selected for the study is the diagnosis and factors affecting Chronic Kidney Disease. The dataset used for the study consists of 153 cases and 11 attributes of CKD patients. The objective of this research is to compare the performance of Artificial Neural Networks (ANNs) and Logistic Regression (LR) classifier on the basis of  the following criteria: Accuracy, Sensitivity, Specificity, Prevalence, and Area under curve (ROC)  for CKD prediction.  From the experimental results, it is observed that the performance of ANNs classifier is better than the Logistic Regression model. With the accuracy of 84.44%, sensitivity of 84.21%, specificity of 84.61% and AUC_ROC of 84.41%. Also, through the final fitted models used, the most important factors that have a clear impact on chronic kidney disease patients are creatinine and urea.