Investment Thesis

This strategy is fundamentally based on the Momentum Effect in Stocks strategy, enhanced by Monte Carlo simulation to optimize portfolio allocation, with additional mechanisms for take-profit, stop-loss, and quick rebalancing.

The momentum anomaly posits that assets with strong recent performance are likely to continue performing well in the near future. Specifically, stocks that have outperformed their peers over a 3-12 month period tend to maintain their upward trajectory. This strategy aims to capitalize on the momentum effect by focusing on large-cap stocks.

The universe selection API constructs a momentum portfolio by filtering out stocks priced below $5 and excluding ETFs without fundamental data. It further narrows the selection to the 70 largest companies based on market capitalization. The momentum percent, representing the relative difference in stock prices over time, is calculated using the formula \(\frac{c_t - c_{t-n}}{c_{t-n}}\), where \(c_t\) is the closing price on day \(t\) and \(c_{t-n}\) is the closing price \(n\) days prior. The algorithm tracks this momentum metric for each symbol using a custom momentum tracking mechanism, with daily updates made using the latest closing prices. A 12-month period is used for calculating momentum, and stocks exhibiting the best 12-month momentum are equally weighted in the portfolio.

In addition to this momentum strategy, a Minimum Variance Portfolio (MVP) optimization approach is applied to find an optimal asset allocation based on a Monte Carlo simulation. This optimization focuses on a shorter, 63-day lookback period to capture more recent performance trends. The Monte Carlo simulation runs 1000 iterations to find a suboptimal point, balancing return and risk. The Sortino ratio, which considers downside risk, is used as the performance metric for optimization.

The portfolio is managed through a combination of monthly rebalancing and dynamic event-driven adjustments:

1. Selection Based on Momentum:
  - The universe is filtered to the 200 most liquid stocks.
  - From these, the top 70 by market capitalization are selected.
  - A momentum score is calculated for each of these stocks, and the top 5 stocks are chosen based on their momentum.

2. Optimization and Adjustment:
  - A Monte Carlo simulation is conducted to find the optimal weights for the selected 5 stocks, considering only the past 63 days of returns.
  - The portfolio is adjusted to these optimal weights, ensuring that the allocation is dynamically adapted based on recent market data.

3. Take-Profit and Stop-Loss Mechanisms:
  - A take-profit event is triggered if the portfolio's profit exceeds 10%, followed by a subsequent drop of 10% from the highest achieved profit. Upon triggering, the portfolio is liquidated, and a quick rebalancing occurs immediately after.
  - A stop-loss event is triggered if the portfolio experiences a loss of 12% or more since the last adjustment. In this case, the portfolio is liquidated, and trading may be temporarily paused until the next scheduled adjustment.

4. Quick Rebalancing:
  - After a take-profit event, the strategy allows for immediate rebalancing to capture new momentum opportunities. This quick adjustment ensures that the portfolio remains aligned with the current market dynamics, potentially maximizing gains while protecting profits.

The combination of momentum-driven stock selection, optimized portfolio allocation, and dynamic event-driven adjustments provides a robust strategy designed to capitalize on market trends while managing risk.