Congratulations on making it through this lesson. We reviewed some foundational building blocks that you will see again and use when we translate academic papers into alpha ideas and then implement those ideas in code. But let’s just remember that, even after testing, if your alphas look promising, this is really only the first stage of their lives so to speak. Lets step back a bit and see how the search for alpha factors fits into the big picture of the quant workflow. We first propose and generate alpha factors, then evaluate them to find those that might show some promise. Then we perform out-of-sample testing of the alpha factors using historical data that wasn’t used to construct the alpha vector. If that looks promising, then we would conduct paper trading in which we don’t use real money, but we follow the factor as if we’re making theoretical trays over time on newly arriving live market data for some period. If that showed promise, then we would put the alpha into production in a real portfolio with real money. The alpha at that stage would be blended with other alphas and the final alpha vector would pass through a portfolio optimizer. We would likely start by giving that alpha factor a small weight in the combined vector, and given more weight if it performed and improved portfolio performance. We would monitor the alpha factor over time knowing that at some point, the factor’s usefulness will erode because we’re trading in a competitive market. Then, we would remove the alpha factor and go back to the beginning to search for new promising alpha factors. Note that when I say “we” in those steps, after the first one, the “we” could be a human portfolio manager or it could be an artificially intelligent agent and AI. Even in the first step, we can utilize AI techniques to help us recover both these AI applications. The application for alpha generation and the AI application for portfolio construction in term two of the API for trading nano degree program. We will cover later stages of this process in future lessons. We will cover backtesting, out-of-sample testing and other methods to avoid overfitting. It can’t be overstated how important out-of-sample testing is in practice. If our goal is to seek above average performance in real production trading with real money and real world constraints, then it is critically important to avoid overfitting to pass data to the point that our alpha factors cannot generalize to current market conditions. Even within the academic scientific community, both in finance and in the general sciences, there are serious discussions about overfitting and about validating the results of research. The rest of this lesson will focus on generating and evaluating alpha factors. But I do hope you keep in mind that it’s perfectly normal if you feel the need to go back and review the concepts covered in this lesson. I know there was quite a lot of material, but I’m also very excited for you as you’re learning the techniques and the type of thinking that real quant researchers deploy in their search for alpha. See you in the next lesson.