Group Method of Data Handling model using modified Levenberg-Marquardt technique


Roudbaneh, Maryam Pournasir (2013) Group Method of Data Handling model using modified Levenberg-Marquardt technique. PhD thesis, Multimedia University.

Full text not available from this repository.


System identification techniques are used in order to model and predict the behaviors of unknown and very complex systems based on known input-output data. A major difficulty in modeling complex systems is the researcher introducing their own presumptions into the model. Group Method of Data Handling (GMDH) algorithm is a self organizing approach and introduced by A. G. Ivakhnenko that allows the researcher to build models of complex systems without making assumptions about the internal workings. In recent years, Application of GMDH algorithm has led to a great success in a broad range of areas such as engineering, science and management. Choosing an appropriate topology has a great impact on the model performance. In addition, it is crucial to calculate the coefficients correctly, which can be rather difficult and time-consuming. Conventional GMDH uses a statistical method called the “linear least squares regression” to calculate the values of the coefficients.

Item Type: Thesis (PhD)
Additional Information: Call No.: QA402 M37 2013
Uncontrolled Keywords: System analysis--Data processing (GMDH algorithms)
Subjects: Q Science > QA Mathematics > QA299.6-433 Analysis
Divisions: Faculty of Computing and Informatics (FCI)
Depositing User: Ms Nurul Iqtiani Ahmad
Date Deposited: 12 Jan 2016 11:09
Last Modified: 12 Jan 2016 11:09


Downloads per month over past year

View ItemEdit (login required)