Applied Nonparametric Regression Methods as Alternatives to the Linear Regression Methods (Comparative Study)

Linear regression methods impose strong constraints on regression models, especially on the error terms where it assumes that it is independent and follows normal distribution, and this may not be satisfied in many studies, leading to bias that cannot be ignored from the actual model, which affects the credibility of the study. We present in this paper the problem of estimating the regression function using the Nadarya Watson kernel and k- nearest neighbor estimators as alternatives to the parametric linear regression estimators through a simulation study on an imposed model, where we conducted a comparative study between these methods using the statistical programming language R in order to know the best of these estimations. Where the mean squares errors (MSE) was used to determine the best estimate. The results of the simulation study also indicate the effectiveness and efficiency of the nonparametric in the representation of the regression function as compared to linear regression estimators, and indicate the convergence of the performance of these two estimates.

Solving the problem of multi collinearity in regression model

In this research ,we studied the problem of multicollinearity among independent variables in the multiple regression model this matter leads to a mistake in one of the essential conditions of the multiple regression model and getting incorrect results. At the beginning we have introduced documented theoretical study of the kinds of the multicollinearity and of the reasons of the problem of the multiple regression model and some methods to discover them. In addition to this we mentioned some methods that treat the cases of multiple regression model then we introduced a new method to treat multicollineartiy and apply it to an example . In this method we have dealt with multicollinearity on the hand and solved the problem of discrepancy between the significant of the regression model and the non-significant of one or more coefficient.

The Effect of Both the Size and The Type of Statistical Sampling on The Estimates of Simple Linear Regression Equation Coefficients

These papers aim to study the estimation of the simple linear regression equation coefficients using the least square method at different sample sizes and different sampling methods. And so on, the main goal of this research is to try to determine the optimum size and the best sampling method for these coefficients. We used experimental data for a population consist of 2000 students from different schools all over the country. We had changed the sample size each time and calculate the coefficients and then compare these coefficients for different sample sizes with their coefficients of the real population; and the results have been shown that the estimation of the linear regression equation coefficients are close from the real values of the coefficients of the regression line equation for the population when the sample size closes the value (325). As it turns out that the Stratified random sampling with proportional distribution with class sizes gives the best and most accurate results to estimate linear regression equation with least square method.

Solving the problem 0f discrepancy between the significance of regression model and non- significance of one of its parameter

The problem of outliers values is considered one of the important objects that affect the estimation of statistical intermediaries specially regression model intermediaries. We have tried in this research to introduce a comprehensive study Of the outliers and some ways of exploring and treating them. In addition to , we have put anew algorithm that helps to find the estimation of the outliers. This matter helps in treating outliers and getting exact results in regression analysis.

ONE: Toward ONE model, ONE algorithm, ONE corpus dedicated to sentiment analysis of Arabic/Arabizi and its dialects

Arabic is the official language of 22 countries, spoken by more than 400 million speakers. Each one of this country use at least on dialect for daily life conversation. Then, Arabic has at least 22 dialects. Each dialect can be written in Arabic or Arabizi Scripts. The most recent researches focus on constructing a language model and a training corpus for each dialect, in each script. Following this technique means constructing 46 different resources (by including the Modern Standard Arabic, MSA) for handling only one language. In this paper, we extract ONE corpus, and we propose ONE algorithm to automatically construct ONE training corpus using ONE classification model architecture for sentiment analysis MSA and different dialects. After manually reviewing the training corpus, the obtained results outperform all the research literature results for the targeted test corpora.