Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
Momentum Driven Factor using Z-score

This post discusses a momentum based strategy to identify potential opportunities when the ticker has forward momentum.

The factor under consideration is a variant of the simple moving average crossover. Two simple moving averages were consider, the 50 day moving average and the 200 day moving average. The ratio of the 50-sma to the 200-sma is then normalized using the z-score. The general idea behind this metric is that the greater the divergence between the 50-sma and the 200-sma, the stronger the signal that the ticker is moving in a certain direction . A positive divergence indicates that the ticker has forward motion. While a negative divergence indicates that the stock has downward motion.

One flaw in the factor is that a ticker can have a positive divergence with downward motion and a negative divergence with forward motion. Refining the factor will need further study. Several avenues exist for additional study that could possibly improve the predictability of the factor. These include:
1. Evaluating different combinations of the moving averages
2. Replacing the simple moving average with the exponential moving average
3. Segmenting the tickers by float, marketcap and/or sector
4. Assessing the optimal number of quantiles