Similarity of distributions which belongs to data estimated by regression analysis and real data

Regression analysis is a statistical method widely used in the literature. One of the most common uses of this method is to make a prediction. However, some assumptions must be satisfied in order to use this method properly. Each regression model obtained may not be statistically valid. Models need...

Full description

Saved in:
Bibliographic Details
Published in2017 Medical Technologies National Congress (TIPTEKNO) pp. 1 - 4
Main Authors Zihni, Nihat Burak, Uzun, Ozlem, Yeginoglu, Gulay, Kalkisim, Sahi Nur, Oksuz, Canan Erthemoglu
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.10.2017
Subjects
Analytical models
Data models
model adequacy
Nanoelectromechanical systems
Predictive models
Qualifications
Regression analysis
similarity of two distributions
Online AccessGet full text

Cover

More Information
Summary:Regression analysis is a statistical method widely used in the literature. One of the most common uses of this method is to make a prediction. However, some assumptions must be satisfied in order to use this method properly. Each regression model obtained may not be statistically valid. Models need to meet certain qualifications in order to be valid. There are various regulatory model qualification conditions that are accepted in the literature. In this study, it is aimed to determine whether the similarity of the distribution of the actual data with the sample of the predicted values of a regression model which is known to be sufficient, that is, whether two data sets belong to the same population. In other words, the population-sample relation that forms the basis of statistical analysis can be established between predictive values obtained by regression model and actual data by various statistical tests.
DOI:10.1109/TIPTEKNO.2017.8238065