IRAQI JOURNAL OF STATISTICAL SCIENCES

The problem of finding genetic mutations is an issue of special importance for its relationship with genetics evolution of living organisms. In this paper we present three statistical methods proposed for the creation of genetic mutations, the first depends mainly on the Markovian model، the second on the autoregressive models, and the third depends on the algorithm proposed based on the hidden Markov model. The three methods are applied to DNA for sequence with shortage oxygen of humans, and shows the practical application show that of these methods give good results in the detection of genetic mutations, reached in some cases to 90%.

In this paper eigenvalue of a gray image is used for image segmentation by calculating the eigenvalue for each segment and to be compared with a threshold value [which will be calculated dynamically by the system] and an iteration procedure will continue running until saturation arrived.
By using the quadtree technique the gray image is divide into small segments depending on the value of the threshold and then the segmented image will pass into low pass filter in order to produce more enhancement in the image .
The algorithm was applied on different types of images [natural images , text images] and both of them give high quality of segmentation .

The non-linear conjugate gradient method is a very useful technique for solving Large-Scale minimization problems and has wide applications in many fields . In this paper, we present a new spectral type, a non-linear conjugate gradient algorithm the derivatation of this algorithm is based on Fletcher – Reeves and Newton algorithm, the descent property for the suggested algorithm is proved provided that the step size satisfies the Wolfe conditions. Numerical results show that the new algorithm is efficient in practical computation and superior to the Fletcher – Reervs algorithm in many situations.

This Paper deals with the recursive identification problem of stochastic linear dynamical systems , Important Algorithms are explained in the identification system domains that are time-varying, and using a recursive Least Square method with a famous approach to estimate the model parameter, that a forgetting factor, and a Kalman filter approach with different values for a best linear dynamic models, that are identified from the two type of stochastic linear dynamic systems: that the equation error models which contain ARX models and ARMAX models. Output error models which consist of OE and Box-Jenkins models, that are reached by using the suggested instrument in Off-Line Identification, where the exact Linear models reached their parameter stable with Time, Moreover, the Statistic terms are verified from a point of random errors and insignificant cross-correlation between inputs and outputs residual.

In this research the covariance matrix which was used in so many fields, its eigen values adopted to be the main parameters for Latin printed character recognition.
The idea was divided in two stages. The first stage is to generate the covariance matrix then, evaluate its eigen values to build main table for the whole latin characters in addition to the numeric values. The second stage is the recognition stage, which achieved to any character that was entered.
The applied examples do not register any negative result. So it can be strongly recommended for printed character recognition.

This paper deals with the problem of estimating the order of Markov chains. An algorithm is proposed for this purpose and programmed by MATLABR2009a, such that the maximum order is selected automatically after inputting of observations and the upper limit of the order. A simulation study is carried out to verify the efficiency of the algorithm, then the algorithm is applied to the observations of rainfall in the city of Mosul for the years 1994-2000 .It is clear from this application that the daily rainfall in Mosul city is with order 1.

In this research we display some information criteria for time series like AIC,BIC,AICc and others. Beside that a proposal criteria is NAICc , two methods have been used for comparison of performance criteria by simulated and real data and we got encouraging results.

This work deals with descriptive summary of linear programming subject, and describes least square method for estimating regression parameters. As well as using other methods in order to estimate the parameters unconventional on reduced variance (or reduce sum of square error) , but depend on minimizing variation of absolute values from the median , the aim of this work is to find an easy and precise way to estimate the absolute deviation , which is important when the variance fail in precise estimate and lead to enlarge the variation of the data , while least square method depend on minimizing sum of the square , therefore , the goal is to reduce the deviation of the absolute values from the mean in model of linear programming in order to estimate absolute deviation which is a simple and precise method , and this manuscript includes examples of this application.

In this research a Distributed Computer System Application for Traffic office has been designed, by using two models of building Distributed Systems which are: Client/Server Model and File Transfer Model, and Two organizations for designing Distributed Databases which are: Central Database with multiple access from several sites , and the organization of Partially replicated databases.
We Designed a Multimedia Database and used it to store the required data including images of vehicles and their owners in tables of the Database, with the ability to extract Statistical Information and display it in the form of tables and charts.
A database driven web site has been created, using the concepts of Web-Databases; the site provides the ability to query for the data of vehicles and their owners in other provinces.
The designed system is homogenous and uses computer devices of type (Pentium IV) as hardware and (Windows) operating system as environment.VFP 9 Language has been used for programming the application, and also as DBMS, the web pages have been Designed using ASP technique, and it's code written using VBscript and HTML Languages.

In this research, genetic algorithm is used to solve Queuing problem, which is considered to be of special importance as a result of the costs emanating from waiting and operating, is suggested. The problem aims at locating the period of time for waiting on the distant range and making this period of time as less as possible .The application of this genetic algorithm to this problem led to limiting the most ideal number of the people served and making the total cost concerning the party who offers the service as less as possible during the period of time specified for waiting.

In this research we study the fuzzy nearest neighbor method (FNNM) for time-series prediction , this method depends on fuzzy membership value .The main goal of the prediction algorithm is to forecast future value depending on past nearest neighbors value. The nearest neighbors that we choose by using fuzzy membership value or portmanteau membership value .
To measures the accuracy of our method and to compare with ARIMA model we use mean absolute percentage error (MAPE) and mean square error (MSE) that is calculated from the actual value of time series data and the number of internet users and forecasting value. The results encourage using fuzzy nearest neighbor in forecasting.

Is in this research the ordinary least squares method is reviewed and a brief account on the multiple linear regression model and regression quintile is given and then study the characteristics of the society and extracting information from them to add to the information data to get the multiple regression model which has less error than original after the addition of information of population, using regression quantile and comparison between the results using criterion Mse.

The main objective of this study is to select the best multiplicative seasonal model of the time series data using some information criteria; they are AIC, BIC, AICC, and SMAIC. Then prediction is made for the different models and calculates the final prediction error criterion to ensure the accuracy of our best selected model. In our study we select the model ARIMA(0,1,1)(0,1,1)12 as the best model for the wind speed series and the model ARIMA(0,1,1)(0,1,1)12 as the best model for the cloud series .

In this research the relation among rain , temperature and relative humidity in Mosul city of the time 1988-2001 was determined by applying Granger test to determine and measure the direction of the causative relation among the variables of the research. To make clear whether the chain is stable or not there was a need for using units root test to check the characters of time chains of the sample variables to know the extent of their stability and determining the degree of integrity of each variable alone. The results showed that the variables were highly integrated so , due to that the joint integrity test was used to determine he existence or the absence of the longterm relation among the variables under the study. The analysis showed the absence of the joint integrity among the variables of the research, besides the instability of remains of the joint integrity equation .Also , passage analysis was used to show the causative relation among the variables under the study and this test boosted by results we came up with by applying Granger test.

In this research we use path analysis to study the variables affecting the scientific level of institute students. Data were from the Technical Institute of Nineveh Examination Committee for students of the first stage Department of Financial and Banking for the academic year 2008-2009 which contains eight independents variables represent study materials and dependent variable which represents of student final average and the sample size is (76) students, and then path analysis is applied by configuring the proposed causal models and then used Stepwise Regression method for a sample estimator for each proposed model using the SPSS Ver. (17) and the program AMOS Ver (5) to find the of direct and indirect effects and total effects for all estimated models, as well as charting the nets of all models using the AMOS results and through the clear importance of the variables first X1 (Internal Banking Operation) , followed by the third variable X3 (Financial Economic) followed by the second variable X2 (Financial Reading) , and finally the fifth variable X5 (Risk Management) in effect on the variable Y (the average of student final).

The use of the Excel program is not only limited to the usual functions and to formulate and solve linear programming models, but beyond that it can be used in the applications of operations research, such as transport problems of the qualities of random in the needs dialing locations in some cases because they are not known for sure, especially when accompanied by imposition of additional constraints for the supplier to impose sanctions when processing more or less than needed.

This paper reviews the architecture of the mathematical model of the problem of transport and indicate the method of random use Excel in a possible solution to the problem to reach the optimal solution quickly and accurately through the property Solver in this program, where the proposed method has been applied to several different examples in the numbers of offer sites and demand sites.

This paper aims to display the concept of beta regression and modality of its utilization in the statistical application by considering the values of the response variable which are constrained by the interval(0,1) and the response variable is following beta distribution. Both Akaike and Bayesian information criteria are used to get the preferable model by applying the multiple beta regression in investigating the percentage of the sugar proportion of the resultant humid spoil to the sugar proportion of the resultant crude juice. We used statistical software R to obtain the expected results.