Abstract: Bank managements are interested in determining the rate of liquidity based on a strategy for meeting this need. Lack of sufficient liquidity might impose heavy costs and even lead to bankruptcy. On the other hand, surplus liquidity will result in losing investment opportunities and reduction in productivity and profit ability of banks. In this research, efforts are taken to use the most affecting factors by investigation on important factors effective on the liquidity rate. The data used are chosen from 573 real values of banking operations. The variables are NEII , RTGS and clear transaction and the outcome of these variables is bank liquidity. In this research, two methods of ARIMA and multiple regressions are applied for predicting the future liquidity data and two mathematical models are introduced. The model introduced can predict liquidity rate based on two years data effectively. The results showed that the mean deviations between output of model and actual results are about 9% and 2.46% in first and second year respectively. This means the people preferred to use the internet in second year more than first year.