A Comparison of Classification/Regression Trees and Logistic Regression in Failure Models

Assessment of the Accuracy of Analyzers for Automatic Determination of PM10 and PM2.5 Particulate Matter in Ambient Air

“…1)the variability condition: to be in line with the uncertainty established by the environmental regulations in force; 2)the Pearson correlation coefficient values, r ≥ 0.97 according to SR EN 14793: 2017 [5]; Pearson correlation and linear regression methods have been used to verify these requirements, these methods being more and more used in different areas of activity, including environmental protection [6][7][8][9].…”

“…Precision is the general term for the variability of the results of a repeated measurement and is usually expressed based on the values of standard deviations obtained under repeatability/reproducibility conditions with a probability of 95%.The paper presents the procedure and the results obtained within a project that aimed the establish the accuracy of measurements of PM2.5 and PM10 made with uRADMonitor A3 a fixed air quality monitoring station ( fig. 1a); the trueness and variability were calculated, based on experimental data obtained by parallel measurement of the concentration of PM2.5 and PM10 using automatic monitors and the reference method, SR EN 12341:2014 and compared with the requirements of acceptability imposed by the regulations under these conditions:1)the variability condition: to be in line with the uncertainty established by the environmental regulations in force; 2)the Pearson correlation coefficient values, r ≥ 0.97 according to SR EN 14793: 2017 [5]; Pearson correlation and linear regression methods have been used to verify these requirements, these methods being more and more used in different areas of activity, including environmental protection [6][7][8][9].…”

Application of MARSplines Method for Failure Rate Prediction

“…the failure rate. In recent years predictive methods, such as the support vector method [13], the K-nearest neighbours method [14], the regression and classification trees method [15] and artificial neural networks [16], have become popular in the modelling of various broadly understood engineering problems. A method exploiting the "splining" of many functions, called the MARS (Multivariate Adaptive Regression Splines) method, is one of the regression algorithms which can be applied to solve variables prediction problems not easily described by typical mathematical models [17].…”

Binary logistic regression—Instrument for assessing museum indoor air impact on exhibits

“…The applications of LR could be found in the economic-financial field, too. Irimia-Dieguez et al (2015) have developed a mathematical model for the economicfinancial analysis of the small and medium enterprises starting from nonfinancial and macroeconomic variables.…”