Cryptocurrency Prediction Assistance

A cryptocurrency bot which can assist you in predicting future crypto prices through comparing previous crypto prices with actual prices. The bot uses the method of Death Cross vs Golden Cross, a reliable method for solid predictions of prices, to make a plot based on previously predicted data.

Image: Preview of the graph

Table of Contents

Features

  • Interactive graph

  • Allows user to input to customize the following in the chart – currency, cryptocurrency, time period and time interval

  • Various built-in chart functions such as zooming and downloading the plotted graph, thanks to plotly

  • Data is viewable for every candlestick in the data chart

  • Along with the graph, a table is also rendered to output important data

Head to project screenshots to check out the mentioned features!

Installation

  • Clone the repository by running the following command
git clone https://github.com/SrNightmare09/crypto-prediction.git
  • Run the following commands to install the required libraries for the code –
pip install yfinance

pip install plotly

  • Run index.py from the src folder. On doing this, A few questions will be asked in your terminal related to the plotting, please give valid input for proper functioning of the graph.

  • When all input values have been received, your terminal will print a table with statistics, along with a plotted graph also being rendered.

  • And, we’re done! But that’s not it, feel free to edit the code as you wish, add features and feel free to send a pull request. Read about contributing here

Note: Currently the project can only be run through raw cloning of the code and running it

How It Works – The Gold Line vs Death Line Algorithm

  • The program receives data of the chosen cryptocurrency from Yahoo! finance, from which it creates a chart based on the original prices of the cryptocurrency.

  • The moving average of 5 days (MA₅) is taken and then plotted in the graph as the ‘Gold Line’, where

MA₅ = (MA₁ + MA₂ + MA₃ + MA₄ + MA₅) / 5
  • Another moving average, this time for 20 days (MA₂₀), is plotted in the graph as the ‘Death Line’, where
MA₂₀ = (MA₁ + MA₂ + MA₃ + MA₄ ... MA₂₀) / 20

  • The graph helps detecting bullish and bearish trends, i.e., helps determine when you should buy or sell. It is explained in the following points.

  • The graph tries to catch a global trend, over which a short-term ‘Gold Line’ and a long-term ‘Black Line’ is plotted

  • The Gold Line will always be more accurate with the prices as it is a short-term moving average compared to the Death Line, which is a long-term moving average

  • Two situations will always be visible in the graph – Death Line being higher than the Gold Line and vice-versa; both these situations will help us make decisions for either buying or selling

  • When the Death line crosses the Gold line and gets higher in the graph, it indicates a bearish moving average in the future market. This situation is known as the Death Cross and it is the perfect time for selling

  • When the Gold line crosses the Death line and gets higher in the graph, it indicates a bullish moving average in the future market. This situation is known as the Golden Cross and it is the perfect time for buying

Read more about Death Cross and Golden Cross here

Resources

Here is a list of resources which helped me, both – the libraries I used and the sources which helped me for the understanding of the algorithm

Libraries used

Sources for understanding the algorithm

I would like to thank for the availibility of these resources as the project would not have been in existence without them!

Upcoming Features

A list of a few features which I will add in the later versions of the project –

  • Error handling for user input
  • Viewing table alongside graph instead of in the terminal
  • Future price predictions
  • Response of code for predicting price by taking real-life events into consideration as well
  • Implement code into a downloadable and useable application
  • Add more graph tools (like drawing tools)
  • Implement shortcuts for graph tools
  • Add dark mode/light mode toggle

Contributing

Contributions are always welcome! Feel free to fork this project, report issues/bugs, etc.

See CONTRIBUTING.md for ways to get started.

Please adhere to this project’s CODE OF CONDUCT.

License

MIT

GitHub

View Github