Stocks are the share of ownership of a company. Stock trading is another avenue through which individuals can make money. Many investors want to buy shares but the prices of these stocks are unpredictable There can be sudden downward trend in stock prices which can result in huge loss of capital. This can discourage the potential investors from buying shares. As a result buyers need intelligent systems that can guide them in buying and selling appropriate stocks at any time. Many intelligent models have been developed towards this end, but their performances can still be improved upon using hybrid models. This paper presents the development of an adaptive neuro-fuzzy model for predicting the values of stock prices so that buyers, especially, short time operators can know which share to acquire or sell at the appropriate time. First Bank stock market prices collected for the period of leven years were used in this research to test the model. The ANFIS model was developed using Matlab. The model is first trained using some of the data points called training data. The model is then tested using the remaining data points called the testing data. The performance of the model on the testing data is 98.01% which shows that the model is suitable in predicting stock data. A comparative analysis of this model with some other classifiers Artificial Neural network (ANN) and Regression Trees in terms of classification accuracy shows a better performance. This work will be very useful for investors making decisions on acquiring and selling their stocks at the appropriate time and will help in minimizing losses.