Kong Yushou, Ji Lingling, Wang Changyu, Li Liguo and Zeng Liming
To forecast the path of tropic cyclones by using a non‐linear statistical forecasting technique – the method of successive analogy.
Abstract
Purpose
To forecast the path of tropic cyclones by using a non‐linear statistical forecasting technique – the method of successive analogy.
Design/methodology/approach
Non‐linear statistical forecasting models can describe the non‐linear relationship between the factors and the forecasting objects and the real atmospheric movement more accurately, so they usually have stronger forecasting capability. In practice, however, it is shown that the relationships between predictors and predictands sometimes are so complex that it is very difficult or even impossible to establish the kind of non‐linear mathematical model. Therefore, it is an important topic for atmospheric science to solve non‐linear prediction problem of atmospheric systems by using the non‐function model approach.
Findings
The objective and quantitative prediction of tropical cyclone moving path can be given by using the method of successive analogy, a non‐linear forecasting technique, and calculating the similarity parameters between the grayscale field and the height field.
Research limitations/implications
Further experiments are needed to verify this technique.
Practical implications
A very useful technique for solving non‐linear problem.
Originality/value
Illustrates the new technique of solving non‐linear statistic problem and its application.