AI Strategies

  • Introduction

    Artificial Intelligence (AI) is a vague concept. In our case, "AI Strategy" is an umbrella term we use to name a set of underlying tools (like machine learning, LLM-based feature engineering, model crossbreeding) which work together so traders could benefit from advances in the AI/ML domain. On ...
  • How AI strategies are different from algorithmic strategies

    Trading strategies originate from some kind of the past experience accompanied with ideas for the future. Let's think of the challenge of building a strategy in the following terms Data. There's a list of historical candles. For each candle, we already know how the price has changed after. Label...
  • Creating an AI strategy: 1. Learning data set

    Models need to be trained on historical data. You should define a market (or a set of markets) for your model to train at. “Market” is a combination of a symbol, chart time frame (like D) and time range (like from date A to date B). You can use any symbols TrendSpider supports, including your own...
  • Creating an AI strategy: 2. What you want to predict

    Remember you’re building a fixed RR strategy. The AI Model will aim to provide a signal when at any time over the course of the next X candles the price will be likely to hit your TP and not your SL. Here’s a picture Remember that the actual trained model won’t catch all the occasions like tha...
  • Creating an AI strategy: 3. Type of the model

    The world of machine learning is not new, and there are numerous machine learning model types out there. Different models work in a different way. There are a few heuristics which can help you to select the model type for the case. However, a typical approach is beautiful and simple: for every ta...
  • Creating an AI strategy: 4. Model inputs (features)

    Defining your model inputs is a make-it-or-break-it kind of a thing. With awesome inputs (that is, "inputs legitimately having predictive power for the signal"), even the simplest model will learn well and work well. With poor inputs (that is, "inputs having nothing to do with the actual signal")...