궤도틀림 진전 예측을 위한 베이지안 분석 방법론 연구

This study is about how to estimate the deterioration progress rate of track quality index representing track geometric irregularities. Most of existing studies used a simple linear regression and regarded the slope of the regression equation as the progress rate. And their model is lack of statistical test like F-value or t-value. In this paper, we present a Bayesian approach to predict the track irregularity progress since the approach is better than simple regression in that the Bayesian model can deal with the prior distribution of parameters and the progress rate could be expressed as a probability distribution. And then we compare two results from simple regression model and Bayesian regression model.