In this chapter, we covered the use of the zipline library for the event-driven simulation of a trading algorithm, both offline and on the Quantopian online platform. We have illustrated the design and evaluation of individual alpha factors to derive signals for an algorithmic trading strategy from market, fundamental, and alternative data, and demonstrated a naive way of combining multiple factors. We also introduced the alphalens library that permits the comprehensive evaluation of the predictive performance and trading turnover of signals. 

The portfolio construction process, in turn, takes a broader perspective and aims at the optimal sizing of positions from a risk and return perspective. We will now turn to various strategies to balance risk and returns in a portfolio process. We will also look in more detail at the challenges of backtesting trading strategies on a limited set of historical data and how to address these challenges.