Study on Probability Estimation of Haze in Beijing Based Logistic Regression Model

摘要

The Logistic Regression Model of two categories is used to explore the relationship between haze and season, various meteorological factors such as air pressure, temperature, relative humidity, precipitation, wind direction and so on. Among all the factors, the relative humidity is best related to haze and season is in the second place. The odds of haze in winter are 17.87 times bigger than that in summer, 3.99 times bigger than that in spring. The odds of haze would increase by 48 percent averagely when the relative humidity increase by 10 percent.