The estimation of the regular regression model requires several assumptions to be satisfied such as "linearity". One problem occurs by partitioning the regression curve into two (or more) parts and then joining them by threshold point(s). This situation is regarded as a linearity violation of regression. Therefore, the multiphase regression model is received increasing attention as an alternative approach which describes the changing of the behavior of the phenomenon through threshold point estimation. Maximum likelihood estimator "MLE" has been used in both model and threshold point estimations. However, MLE is not resistant against violations such as outliers' existence or in case of the heavy-tailed error distribution. The main goal of this paper is to suggest a new hybrid estimator obtained by an ad-hoc algorithm which relies on data driven strategy that overcomes outliers. While the minor goal is to introduce a new employment of an unweighted estimation method named "winsorization" which is a good method to get robustness in regression estimation via special technique to reduce the effect of the outliers. Another specific contribution in this paper is to suggest employing "Kernel" function as a new weight (in the scope of the researcher's knowledge).Moreover, two weighted estimations are based on robust weight functions named "Cauchy" and "Talworth". Simulations have been constructed with contamination levels (0%, 5%, and 10%) which associated with sample sizes (n=40,100). Real data application showed the superior performance of the suggested method compared with other methods using RMSE and R2 criteria.
The purpose of this paper is to model and forecast the white oil during the period (2012-2019) using volatility GARCH-class. After showing that squared returns of white oil have a significant long memory in the volatility, the return series based on fractional GARCH models are estimated and forecasted for the mean and volatility by quasi maximum likelihood QML as a traditional method. While the competition includes machine learning approaches using Support Vector Regression (SVR). Results showed that the best appropriate model among many other models to forecast the volatility, depending on the lowest value of Akaike information criterion and Schwartz information criterion, also the parameters must be significant. In addition, the residuals
... Show MoreTwo groups of chronic hepatitis B and C virus patients were divided into Pre-treated patients (25 CHB patients with positive HBs Ag for more than 6 months and 40 CHC patients), and post-treated patients [12 CHB patients (4, 6, and 2 were treated with lamivudine, IFN-? and combination of LMV + IFN-? respectively), and 27 patients for CHC (3, 13 and 11 patients were treated with Ribavirin, IFN-? and combination therapy (RBV+ IFN-?) respectively].These patients were followed up for 6 months. By using ELISA technique, levels of IL-6, IL-10, IFN-? and TNF-? were measured in vivo and in vitro (supernatant of PBMCs stimulated with PHA) and compared with healthy control. The mean level of IL-6, IL-10 and TNF-? in CHB patients showed significant dif
... Show MoreThe influence of fear on the dynamics of harvested prey-predator model with intra-specific competition is suggested and studied, where the fear effect from the predation causes decreases of growth rate of prey. We suppose that the predator attacks the prey under the Holling type IV functional response. he existence of the solution is investigated and the bounded-ness of the solution is studied too. In addition, the dynamical behavior of the system is established locally and globally. Furthermore, the persistence conditions are investigated. Finally, numerical analysis of the system is carried out.
It is commonly known that Euler-Bernoulli’s thin beam theorem is not applicable whenever a nonlinear distribution of strain/stress occurs, such as in deep beams, or the stress distribution is discontinuous. In order to design the members experiencing such distorted stress regions, the Strut-and-Tie Model (STM) could be utilized. In this paper, experimental investigation of STM technique for three identical small-scale deep beams was conducted. The beams were simply supported and loaded statically with a concentrated load at the mid span of the beams. These deep beams had two symmetrical openings near the application point of loading. Both the deep beam, where the stress distribution cannot be assumed linear, and the ex
... Show MoreThe prediction process of time series for some time-related phenomena, in particular, the autoregressive integrated moving average(ARIMA) models is one of the important topics in the theory of time series analysis in the applied statistics. Perhaps its importance lies in the basic stages in analyzing of the structure or modeling and the conditions that must be provided in the stochastic process. This paper deals with two methods of predicting the first was a special case of autoregressive integrated moving average which is ARIMA (0,1,1) if the value of the parameter equal to zero, then it is called Random Walk model, the second was the exponential weighted moving average (EWMA). It was implemented in the data of the monthly traff
... Show MoreIn this paper, we establish the conditions of the occurrence of the local bifurcations, such as saddle node, transcritical and pitchfork, of all equilibrium points of an eco-epidemiological model consisting of a prey-predator model with SI (susceptible-infected) epidemic diseases in prey population only and a refuge-stage structure in the predators. It is observed that there is a transcritical bifurcation near the axial and free predator equilibrium points, near disease-free equilibrium point is a saddle-node bifurcation and near positive (coexistence) equilibrium point is a saddle-node bifurcation, a transcritical bifurcation and a pitchfork bifurcation. Further investigations for Hopf bifurcation near coexistence equilibrium point
... Show MoreAbstract
The aim of this work is to create a power control system for wind turbines based on fuzzy logic. Three power control loop was considered including: changing the pitch angle of the blade, changing the length of the blade and turning the nacelle. The stochastic law was given for changes and instant inaccurate assessment of wind conditions changes. Two different algorithms were used for fuzzy inference in the control loop, the Mamdani and Larsen algorithms. These two different algorithms are materialized and developed in this study in Matlab-Fuzzy logic toolbox which has been practically implemented using necessary intelligent control system in electrical engineerin
... Show More