Before we begin this article offers no financial advice and is only an experiment carried out using the recently launched large language model from Meta in the form of Llama 2. The explosion of artificial intelligence (AI) in recent years has affected every facet of our life. One of the most intriguing applications of AI in finance is the creation of an AI investment advisor for generating trading strategies.
This experimental guide delves into the process of creating an AI investment advisor thanks to a tutorial video created by Siraj Raval. Its application in predicting stock prices, and the generation of an investment thesis. It also explores the use of Composer for trading strategy creation, backtesting of trading strategies, and the limitations of existing AI models.
AI investment advisors are becoming increasingly popular due to their ability to analyze vast amounts of data and make predictions with a high degree of accuracy. One such AI investment advisor has been developed by Siraj, which generates investment advice and trading strategies. This AI uses Python, a popular programming language known for its simplicity and robustness, to build a time series model. This model is used to predict stock prices, such as Tesla’s next month’s price.
Building an investment advisor with Llama 2
The AI not only predicts stock prices but also generates an investment thesis based on the prediction and the overall market trends. This investment thesis can then be used to create a trading strategy on the trading platform, composer.trade. The trading strategy is typically based on the 50-day moving average price of a company’s stock compared to its 200-day moving average. In backtesting, the AI’s trading strategy has outperformed the SPY by around 700%, demonstrating its potential effectiveness.
Other articles you may find of interest on the subject of Llama 2 :
- Build your own private personal AI using Llama 2
- Llama 1 vs Llama 2 AI architecture compared and tested
- How to install a private Llama 2 AI assistant with local memory
- LLaMA 2 vs Claude 2 vs GPT-4
- Running Llama 2 13B on an Intel ARC GPU, iGPU and CPU
- Llama 2 unrestricted version tested running locally
AI trading strategies
The AI investment advisor was built using an open base model, Llama 2, released by Facebook. This model was fine-tuned on code data (Code Llama), enhancing its ability to analyze and predict stock market trends. The AI can be run on a GPU using a Google Collab notebook with five commands, making it accessible to users with varying levels of technical expertise.
One of the key features of this AI is its ability to pull real-time data from the web to create an analysis and investment thesis. This is achieved through the use of APIs, which allow the AI to access and analyze up-to-date information. The AI outputs code as a response, which is then pre-processed and executed using a Jupyter notebook function. This process ensures that the AI’s predictions and investment advice are based on the most recent data.
Llama 2 investment advisor
While the Llama 2 model has proven effective, other models can also be used. For instance, Replit’s three billion parameter model, Replit Code V1 3B, which was trained on 20 different languages, can also be utilized. However, it’s important to note that running inference on GPUs is not cheap, but it is cheaper than paying OpenAI.
The AI can be used to generate unrestricted investment advice and can be extended to make investments. This makes it a versatile tool for both individual investors and financial institutions. For executing the trading strategies, Composer is recommended as it provides a dashboard and a community of different strategies.
The use of AI in creating investment advisors and generating trading strategies has the potential to revolutionize the financial industry. With the ability to analyze vast amounts of data and make accurate predictions, AI investment advisors can provide valuable insights and advice to investors. However, it’s important to be aware of the limitations of existing AI models and the costs associated with running inference on GPUs. As the technology continues to evolve, it’s likely that we’ll see even more sophisticated and effective AI investment advisors in the future.
Latest Geeky Gadgets Deals
Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.