Deming regression and its application on the derived data

Regression with the simplest definition: It is a statistical method supposed to reveal the dimensions of two variables known to have a relationship between themselves. While the existence of the relation between either dependent or independent two variables is being reviewed with many other methods used in the statistics, an equation could be established by a mathematical interpretation of the relationship made in the Regression Analysis which reveals the affecting ratio of the independent variable on the dependent variable, called a Regression Equation. The results to be revealed can be estimated thanks to this equation without the need of any application, just by changing the value of numbers of the equation, instead of costly trials. Today, once the necessary values of the patients are known, especially the drug doses in the medical sector are being determined by making use of the Regression Analysis. The Regression analysis can be made by several different sub-methods. One of the most commonly used sub-methods is the Least Squares method. The Least Squares method is being used to estimate the average values of the dependent variable in cases where the measured values are evident. The measured values are assumed to be correct and the whole process are being continued based on this assumption. The regression techniques where the assumption that there is no any error in the statistical measurement is being accepted, are called Type I Regression Types. Beside the Type I regression types where there is a probability of measurement errors, there is also a best regression technique that can be used in such cases and it can be said that it is the Deming Regression. The Deming Regression is aiming at achieving the best predictive equation by taking into consideration the measurement errors occurring in the phase of obtaining the independent variables. In this study, 72 X and Y values, which conformed to the normal distribution, have been randomly generated and the relation between the values of the independent variable X and the values of the Dependent Y have been examined by the Deming Regression technique. As the outcome of the processes, beside the errors of Y variable that could be identified in many other regression techniques, the errors of X variable which have measurement values have been also identified and all elements that may influence the result in the model have been taken into consideration. The Deming Regression Technique, beside being able to calculate the errors of the dependent variables, it also calculates the measurement errors in the independent variables and ensures better and more reliable results to be obtained.