Dear Qasim Shah, the Neural network is a data-driven model that mimics the learning process of the biological neural network. The first thing you would need is data set representing the problem you want to solve. For instance, in the case of stock market price prediction; you would need historical data of previous stock market prices as dependent variables and the corresponding values of the independent variables or features i,e. those factors on which stock market price depend on. After getting this data set, you can carryout some pre-processing such as feature normalisation, data cleaning and so on. Then split your data set into training set, validation set and test set. Finally, you can train the neural network by presenting it with this data set. You can consult materials online to assist you further.
Do you have a working knowledge in any programming language? If it is the case which one(s)? Then maybe we can further help you to use some certain technique and framework.
I'm afraid you need to know some programming, for preprocessing and manipulating the data that you'd eventually need to presumably feeding a system of this kind. Or simply you'll ask for it from a friend who actually knows it or simply hire a programmer.