Variables in Pattern Recognition: Machine Learning for Algorithmic Trading in Forex and Stocks p. 13

TL;DR

This content discusses the various explicit and implicit variables involved in pattern recognition and machine learning, highlighting the importance of finding a balance between opportunity and accuracy.

Transcript

you all righty everybody time to gather our thoughts and plant our feet so far we've only done a very very few basic things here so it might seem like we've only just barely left the beach but the reality is if we turn around we realize we're actually in very deep waters already and we are very far from the beach let's consider the current situatio... Read More

Key Insights

• 💱 Explicit variables like percent change, pattern length, and pattern evaluation methods can impact pattern recognition accuracy.
• 🤕 Assigning value or weight to patterns based on their age or freshness can affect prediction performance.
• ⌛ The time frame for outcome evaluation and the required similarity for pattern matches can impact the decision-making process.
• 🤝 The challenge of dealing with an infinite number of possibilities arises when considering the level of decimal accuracy required in pattern evaluation.
• 🪡 The implicit variable of opportunity versus accuracy highlights the need to balance the volume of trade opportunities with prediction correctness.
• ⚾ Adjusting variables based on their performance trend is a common approach in machine learning.
• 🧑‍🏭 Over time, the variables in pattern recognition and machine learning may change due to factors like volatility and the overall economy.

Questions & Answers

Q: What are the explicit variables used in pattern recognition discussed in the content?

The explicit variables discussed are percent change as a means of recognizing patterns and different approaches like point by point and end to beginning percent change.

Q: How does pattern length affect pattern recognition?

Pattern length can vary, but the content suggests a range between ten and a thousand pieces of data. Longer patterns may provide more information but could increase complexity.

Q: How does the value or weight of patterns affect their evaluation?

Currently, the content suggests a fixed value or weight for patterns, regardless of their age or freshness. However, the evaluation of older patterns may differ in terms of their relevancy and accuracy.

Q: What is the implicit variable discussed in the content?

The content discusses the implicit variable of opportunity versus accuracy, which involves finding a balance between the volume of trade opportunities and prediction accuracy.

Summary & Key Takeaways

• The content discusses explicit variables in pattern recognition, such as percent change and pattern length, and explores different approaches like point by point and end to beginning percent change.

• It raises the question of how to assign value or weight to patterns based on their freshness and age, and the potential impact on prediction accuracy.

• The content also touches on considering the time frame for outcome evaluation, the required similarity for patterns to be considered a match, and the challenge of dealing with an infinite number of possibilities with implicit variables.

• It concludes by highlighting the implicit variable of opportunity versus accuracy and the importance of finding a balance between volume of trades and prediction correctness.