Academic Journals Database
Disseminating quality controlled scientific knowledge

Matrix Driven Multivariate Fuzzy Linear Regression Model in Car Sales

Author(s): Lazim Abdullah | Nadia Zakaria

Journal: Journal of Applied Sciences
ISSN 1812-5654

Volume: 12;
Issue: 1;
Start page: 56;
Date: 2012;
VIEW PDF   PDF DOWNLOAD PDF   Download PDF Original page

Keywords: error analysis | coefficient of determination | fuzzy linear regression | Car sales volume | matrix

Fuzzy linear regression has been used in predicting analysis as to handle uncertainty variables. Many methods of fuzzy linear regressions were introduced but most of the methods associated with substantial complex computation procedures. The model of matrix-driven fuzzy linear regression was proposed as to overcome the computational risk and was successfully tested in a civil engineering application. This study extends the application of the model to investigate the relationship between variables impacting car sales volume. The variables of petroleum prices, population, Gross Domestic Product (GDP) and Gross National Product (GNP) are predicted with the response variable of car sales volume. Thirty years time series data of the variables from various Malaysian agencies were fed into the models. It is found that the model successfully yield a fuzzy linear regression equation as to explain the relationship between predictors and response variable. It also notices that eighty eight percent variations in car sales volume attributed by price of petroleum, population, GNP and GDP. The model also successfully explained the contributions of left and right errors of fuzzy numbers of regression coefficients to the car sales volume. The fuzzy numbers that represent coefficients of regression certainly offer a new contribution to the relationships between the variable of car sales volume and the four predictors.
Save time & money - Smart Internet Solutions      Why do you need a reservation system?