jcwf

Journal of Climatology & Weather Forecasting

ISSN - 2332-2594

Abstract

Modeling of Data from Climate Science Using Introductory Statistics

Mahamad B Pathan

The attempt is made for building model(s) to climate data by using introductory statistics. The application of simple statistical methods can expose important insights in climate data. Various statistical tools from introductory statistics help to analyze climate data. The correlation analysis also helps to study interrelations between the variables in given data. The tools such as simple linear regression and multiple regressions are used to fit model for given climate data. Estimation/Prediction of dependent variable is made by using fitted models. The reliability of a model is discussed through residuals. A numerical example consists of three variables such as maximum temperature, minimum temperature and rainfall is considered to illustrate the analysis. The importance of variables in a given data is checked through simple statistical analysis. It is observed that 92% variation in rainfall is explained by range temperature (max-min). The multiple regression models provides better estimate for rainfall rather than two variable analyses.

Top