📦 OpenFintech: Opensource Fintech Library

1. Introduction

OpenFintech is a pioneering project designed with the primary objective of allowing users to create custom high-frequency trading algorithms and introducting them to financial technology as a field 💻. This project also includes wrappers for SQLite3 and the Alphavantage API (a financial information API).

1.1. Who Is OpenFintech For?

OpenFintech is tailored for:

Students who have a keen interest in:
- Delving deeper into trading algorithms.
- Grasping the basics of algorithmic trading and fintech.
- Undertaking both backtesting and simulated live testing of their algorithmic trading strategy.

2. The Heart of OpenFintech: Mean Reversion Strategy

2.1. Conceptualizing Mean Reversion

Mean reversion is the cornerstone of the strategies offered by OpenFintech. It is premised on the theory that asset prices, volatility, and historical returns will eventually revert to the long-run mean of the entire dataset. In the financial realm, this mean can manifest in various contexts, such as a stock's historical price data, P/E ratio or the average return of an industry sector.

2.2. Potential Risks

It's crucial to note that a return to the norm isn't always guaranteed. Unexpected market highs or lows could signal a shift in the usual trend. Events that might trigger such shifts include product launches or, on the flip side, recalls and lawsuits.

3. The Brain Behind the Operations: Model.py and Backtesting

At the core of OpenFintech's algorithmic strategies is the Model.py file, which focuses on backtesting the strategies to provide users with realistic and insightful simulations.

3.1. Algorithmic Trading Strategies our system offers (for now at least 👀)

Moving Average (SMA) with Different Periods:
- In this strategy, the system considers two distinct moving averages: one short-term (like the 50-day moving average) and one long-term (like the 200-day moving average).
- Long Signal: When the short-term moving average crosses above the long-term moving average, this is a potential sign that the asset is in an upward momentum, thus generating a buy (or "long") signal.
- Short Signal: Conversely, when the short-term moving average crosses below the long-term moving average, the asset might be experiencing a downturn, thus generating a sell (or "short") signal.
- Customizable parameters:
  - Moving averages (e.x. SMA and EMA) and their associated periods

4. How To Use

Here's a step-by-step guide to using the OpenFintech package:

Installation: Begin by installing the OpenFintech package using pip.
```
pip install OpenFintech
```
Setting Up The File: After installation, create a Python script and import the required classes from the OpenFintech package.
```
from OpenFintech import Model
```
Variable Configuration: Before proceeding, set up the necessary variables for your database and API configurations. It's recommended to use environment variables to store these values securely.
```
ALPHAVANTAGE_KEY = "RANDOM KEY" # get your own key here https://www.alphavantage.co/support/#api-key
```
Initialization: With the above variables, initiate the various handlers:
- Model handler
```
handler = Model() # or provide a db_name.db
```

Setting Backtesting Parameters: Define a dictionary for backtesting settings. This dictionary will contain parameters like starting amount, short and long strategies, stock ticker, and more.

setting_values = {
    "user_id": 1,
    "starting_aum": 100000,
    "short": "EMA 5",
    "long": "SMA 15",
    "ticker": "NIO",
    "stop_loss": 0.10, #%
    "take_profit": 0.5,#%
    "chart_freq_mins": 0
}

Backtesting: Call the backtest function from the model handler with the previously defined settings and the alphavantage key
```
response = handler.backtest(setting_values, ALPHAVANTAGE_KEY)
print(response)
```
Clean Up: After completing all operations, make sure to close the connection to the database.

   handler.disconnect()
   print("Disconnected database connection.")

Note: Always ensure that sensitive data such as passwords and API keys are stored securely and are not hard-coded directly into the script.

5. Output

The output of Model.backtest contains the price data and the trade data as a pandas df along with the performance of the settings provided to the algorithm.

{'price_data': 1. open 2. high 3. low 4. close 5. volume 6. EMA5 7. SMA15 2023-04-18 187.150 187.69 183.5775 184.31 92067016.0 185.355453 189.188667 2023-04-19 179.100 183.50 177.6500 180.59 125732687.0 183.766969 188.615333 2023-04-20 166.165 169.70 160.5600 162.99 210970819.0 176.841313 186.556000 2023-04-21 164.800 166.00 161.3208 165.08 123538954.0 172.920875 184.542667 2023-04-24 164.650 165.65 158.6100 162.55 140006559.0 169.463917 181.548667 ... ... ... ... ... ... ... ... 2023-08-14 235.700 240.66 233.7500 239.76 98595331.0 244.249615 254.577333 2023-08-15 238.730 240.50 232.6100 232.96 88197599.0 240.486410 252.422667 2023-08-16 228.020 233.97 225.3800 225.60 112484520.0 235.524273 249.839333 2023-08-17 226.060 226.74 218.8300 219.22 120718417.0 230.089516 247.406667 2023-08-18 214.120 217.58 212.3600 215.49 136276584.0 225.223010 244.010000
[86 rows x 7 columns], 'ending_aum': 152490.91942089135, 'dollar_change': 52490.919420891354, 'percent_change': 52.49091942089136}

6. Conclusion

OpenFintech, with its emphasis on the mean reversion strategy, provides an unparalleled platform for budding financial analysts and enthusiasts to test, refine, and understand the nuances of algorithmic trading. Whether you're a student or an expert, OpenFintech's extensive features ensure that you're equipped with the tools needed to delve deep into the world of algorithmic trading.

Laurier-Fintech/OpenFintech