DEVELOPING FACTOR-BASED TRADING STRATEGIES
The development of a trading strategy has many similarities with an engineering project. We begin by designing a framework that is flexible enough so that the components can be easily modified, yet structured enough that we remain focused on our end goal of designing a profitable trading strategy.
Basic Framework and Building Blocks
The typical steps in the development of a trading strategy are:
- Defining a trading idea or investment strategy.
- Developing factors.
- Acquiring and processing data.
- Analyzing the factors.
- Building the strategy.
- Evaluating the strategy.
- Backtesting the strategy.
- Implementing the strategy.
In what follows, we take a closer look at each step.
Defining a Trading Idea or Investment Strategy
A successful trading strategy often starts as an idea based on sound economic intuition, market insight, or the discovery of an anomaly. Background research can be helpful in order to understand what others have tried or implemented in the past.
We distinguish between a trading idea and trading strategy based on the underlying economic motivation. A trading idea has a more short-term horizon often associated with an event or mispricing. A trading strategy has a longer horizon and is frequently based on the exploitation of a premium associated with an anomaly or a characteristic.
Developing Factors
Factors provide building blocks of the model used to build an investment strategy. We introduced a general definition of factors earlier in this chapter. After having established the trading strategy, we move from the economic concepts to the construction of factors that may be able to capture our intuition. In this chapter, we provide a number of examples of factors based on the cross-sectional characteristics of stocks.
Acquiring and Processing Data
A trading strategy relies on accurate and clean data to build factors. There are a number of third-party solutions and databases available for this purpose such as Thomson MarketQA,6 Factset Research Systems,7 and Compustat Xpressfeed.8
Analyzing the Factors
A variety of statistical and econometric techniques must be performed on the data to evaluate the empirical properties of factors. This empirical research is used to understand the risk and return potential of a factor. The analysis is the starting point for building a model of a trading strategy.
Building the Strategy
The model represents a mathematical specification of the trading strategy. There are two important considerations in this specification: The selection of which factors and how these factors are combined. Both considerations need to be motivated by the economic intuition behind the trading strategy. We advise against model specification being strictly data driven because that approach often results in overfitting the model and consequently overestimating forecasting quality of the model.
Evaluating, Backtesting, and Implementing the Strategy
The final step involves assessing the estimation, specification, and forecast quality of the model. This analysis includes examining the goodness of fit (often done in sample), forecasting ability (often done out of sample), and sensitivity and risk characteristics of the model.
We cover the last two steps in greater detail in the companion chapter that follows.