Backtesting, Momentum, and Diversification Lessons from a Quant Internship

Summary

This first-person account describes experiments with a volatility-triggered scalping strategy and a moving-average momentum strategy. The initial scalping rule used Average True Range to set a trading threshold and compared recent price action, but its reported backtest outcomes differed sharply across two historical periods. That contrast illustrates how a strategy can be sensitive to the sample window and may fit particular market conditions. The author then reports applying a moving-average crossover approach to individual stocks and a diversified stock portfolio, where the portfolio results were stronger in the reported tests.

These figures are the author’s backtest observations, not independent validation or evidence of future performance. The account gives little detail on execution assumptions, transaction costs, parameter selection, or out-of-sample testing, so the results are difficult to reproduce. It also mentions sentiment indicators such as volatility measures and put-call ratios, but does not explain a complete strategy for them. The main practical themes are to test across regimes and consider diversification when evaluating a momentum strategy.

Key ideas

• The author used Average True Range to guide a simple volatility-based scalping rule.
• The reported scalping results changed substantially between historical test periods.
• A moving-average crossover strategy was applied to individual stocks and a diversified portfolio.
• The author reports improved portfolio backtest results, though the account lacks enough detail for independent validation.
• Testing across periods and diversifying exposures are emphasized as useful strategy evaluation practices.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.