جستجوی مقالات مرتبط با کلیدواژه « gmdh neural network » در نشریات گروه « مدیریت »

The role of customer relationship management as a strategic tool in development of manufacturing and service organizations, and also acquisition and retention customers in competitive industries, is undeniable. Identification, valuation and classification of customers and allocating resources to them based on their value for organization are the main concerns in customer relationship management. One of the most important tool in this direction, is calculating and predicting customer lifetime value (CLV). “CLV” is a value which is expected customer bring to the organization in specified period.
In this paper, calculating and predicting customer lifetime value is as a key tool in the implementation of customer relationship management in banking. The GMDH neural networks due to its high performance in terms of prediction, is applied and with genuine customer demographic and transactional information of a private Iranian bank , the CLV forecasting is evaluated. The results show that this tool can be used to accurately predict over 90% of customer lifetime value.

This paper aims to find an approach for evaluating and ranking customers, in order to help a bank take effective and strategic decisions in providing services to each customer. Recognition of different ranks of customers based on financial, demographical, and residential features, as effective factors in Customer Lifetime Value (CLV), will help design an appropriate mechanism, which would provide each customer with customized services. In order to calculate the CLV and design such mechanism, the GMDH neural network methodology is used, where both residential and commercial type of demographical and financial variables for customers from two Iranian banks are used as the neural network inputs. Results reveal that financial variables have the most impact on the CLV.

The phenomenon of globalization is one of the most debated issues facing countries in today’s world. There are many different opinions concerning this issue. Investigation of the extensive impacts of this process on different dimensions of human life in political, economic, social and cultural spheres has been preoccupying thinkers, politicians, economists and cultural experts around the world. This study has evaluated various effects of economic globalization on total labor demand broken into skilled and unskilled labor in Iran. In this study we use two models of ARDL and GMDH neural network and also two indexes of foreign direct investment and openness as indicators of globalization for investigating and forecasting the effects of globalization on the labor market. The study based on ARDL approach in the 1353-1385 period shows that globalization indexes have a positive impact on labor demand, but they do not have much impact on labor demand in the short run with this effect becoming increasingly significant in the long run. Also these impacts are greater on the demand of skilled labor than unskilled labor. Finally this study indicates that in all cases, GMDH neural network is a more efficient approach in forecasting employment impacts than ARDL.

This study investigates the impact of globalization on non-oil exports in Iran based on monetary and non-monetary factors through application of GMDH neural network model, as a tool with high ability for tracing and indicating complex nonlinear trends, especially when limited number of observations are available. The research has used International Trade Integration criteria and trade openness index as indices of globalization. The findings of this research illustrate that the impact of the growth of the world income and the growth of imported capital goods on non-oil exports in Iran is more important than the effect of other factors in globalization process. Furthermore, this study concludes the growth of GDP as an internal factor has a significantly major effect on increasing non-oil exports.