FORECASTING MODELS FOR COMMERCIAL BANK ASSET MANAGEMENT: THE CASE OF LIBYA

The purpose of this study is to investigate the banking industry in Libya and to develop some forecasting models in order to help the commercial banks in Libya plan their investment and loan capacity.; Until 1951 Libya was dominated by foreign powers. This foreign domination was one of the main reasons for the country remaining underdeveloped economically and for its banking system remaining less advanced.; Libya is a country with a developing economy where foreign trade plays a major role. The country was in deep poverty until the discovery of oil in the late fifties. With the flow of petroleum revenues development planning began. After 1969, the government intensified its intervention in the economy and by 1979 the government's role in the economy was overwhelmingly predominant in all economic sectors.; Commercial banks in Libya have passed through three stages. The first stage was characterized by foreign domination, the second stage was the Libyanization of the industry, and the third stage was the nationalization of the industry. The major findings were that commercial banks in Libya have had and still have excess reserves that could be utilized and that the country lacks active money and capital markets, which places even more emphasis on the role of the commercial banks.; This study develops some mathematical models based on exponential smoothing, geometric growth, linear regression and naive techniques. The object of these techniques is to develop model(s) that could help commercial banks in Libya forecast the flow of demand, time, savings and total deposits. There were 80 monthly observations used covering the period from January 1971 to August 1977. The data were divided into two parts: base data covering the period from January 1971 to December 1974 and test data covering the period from January 1975 to August 1977. The mean square error (MSE) was used in determining whether one model was superior to other models. A model with the smallest MSE was concluded to be a better forecasting model than those forecasting models with a larger MSE. Six different lags were forecasted; those were one-, two-, three-, four-, five-, and six-month lags.; When applied on the base data, the linear regression technique gave a more accurate prediction than all other models. On the test data, regression technique did not perform as well as in the case of the base data.; The major contribution of this work is that commercial banks in Libya could use some of the models developed in this study that will enable them to forecast the flow of funds. Forecasting deposits will help commercial banks' planning, and this will help them to allocate capital to their most productive sectors.