The GMDH Algorithm of Ivakhnenko
Stanley J. Farlow · The American Statistician · 1981
In 1966, the Russian cyberneticist, A.G. Ivakhnenko, introduced a technique for constructing an extremely high-order regression-type polynomial. The algorithm, the Group Method of Data Handling (GMDH), builds a multinomial of degree in the hundreds, whereas standard multiple regression becomes bogged down in computation and linear dependence. The GMDH method is ideal for complex, unstructured systems where the investigator is only interested in obtaining a high-order input-output relationship. Also, the Ivakhnenko algorithm is heuristic in nature and is not based on a solid foundation as is regression analysis.