Discussing the article: "Reimagining Classic Strategies: Forecasting Higher Highs And Lower Lows"

[MetaQuotes]

Check out the new article: Reimagining Classic Strategies: Forecasting Higher Highs And Lower Lows.

In this series article, we will empirically analyze classic trading strategies to see if we can improve them using AI. In today's discussion, we tried to predict higher highs and lower lows using the Linear Discriminant Analysis model.

Typically, when traders following price action strategies analyze a security, they look for signs of strong trends either forming or dying out. A well-known sign of a strong trend forming is when price levels close above previous extreme values and continue to drift away in progressively larger steps. This is colloquially known as "higher highs" or "lower lows." depending on which direction price is moving.

For countless generations, traders have used this simple strategy to identify entry and exit points. Exit points are generally determined when the price fails to break past its extreme values, indicating that the trend is losing strength and may potentially reverse. Over the years, various minor extensions have been added to this strategy, but its fundamental template remains the same.

One of the biggest drawbacks of this strategy is when the price unexpectedly moves back beneath its extreme. These adverse price changes are known as "retracements" and they are difficult to predict. As a result, most traders do not immediately enter a position when the price breaks to a new extreme. Instead, they wait to see how long the price can sustain those levels before committing—essentially allowing the trend to prove itself. However, this approach raises several questions: How long should one wait before concluding that the trend has proven itself? Conversely, how long is too long before the trend reverses? This is the dilemma faced by price action analysts.

Author: Gamuchirai Zororo Ndawana