AI Trading Model

Introduction

Artificial Intelligence (AI) has revolutionized trading by providing advanced tools to analyze large datasets and make predictions. This project demonstrates how to build a simple AI model for trading using historical price data.

Getting Started

These instructions will help you set up and run the AI trading model on your local machine.

Prerequisites

• Python 3.8 or higher
• pip (Python package installer)
• Jupyter Notebook (optional, for interactive development)

Installation

1. Create a virtual environment:

python <span>-m</span> venv venv
<span>source </span>venv/bin/activate  <span># On Windows use `venv\Scripts\activate`</span>
python <span>-m</span> venv venv
<span>source </span>venv/bin/activate  <span># On Windows use `venv\Scripts\activate`</span>
python -m venv venv
source venv/bin/activate  # On Windows use `venv\Scripts\activate`

Enter fullscreen mode Exit fullscreen mode

Data Preparation

1. Obtain Historical Data:
   Download historical trading data from a reliable source (e.g., Yahoo Finance, Alpha Vantage).

2. Data Preprocessing:
   Clean and preprocess the data to remove any inconsistencies. Typical preprocessing steps include handling missing values, normalizing data, and feature engineering.

Example preprocessing script:

<span>import</span> <span>pandas</span> <span>as</span> <span>pd</span>
<span>from</span> <span>sklearn.preprocessing</span> <span>import</span> <span>MinMaxScaler</span>
<span># Load data </span><span>data</span> <span>=</span> <span>pd</span><span>.</span><span>read_csv</span><span>(</span><span>'</span><span>historical_data.csv</span><span>'</span><span>)</span>
<span># Handle missing values </span><span>data</span> <span>=</span> <span>data</span><span>.</span><span>dropna</span><span>()</span>
<span># Normalize data </span><span>scaler</span> <span>=</span> <span>MinMaxScaler</span><span>()</span>
<span>data</span><span>[[</span><span>'</span><span>Open</span><span>'</span><span>,</span> <span>'</span><span>High</span><span>'</span><span>,</span> <span>'</span><span>Low</span><span>'</span><span>,</span> <span>'</span><span>Close</span><span>'</span><span>,</span> <span>'</span><span>Volume</span><span>'</span><span>]]</span> <span>=</span> <span>scaler</span><span>.</span><span>fit_transform</span><span>(</span><span>data</span><span>[[</span><span>'</span><span>Open</span><span>'</span><span>,</span> <span>'</span><span>High</span><span>'</span><span>,</span> <span>'</span><span>Low</span><span>'</span><span>,</span> <span>'</span><span>Close</span><span>'</span><span>,</span> <span>'</span><span>Volume</span><span>'</span><span>]])</span>
<span># Save preprocessed data </span><span>data</span><span>.</span><span>to_csv</span><span>(</span><span>'</span><span>preprocessed_data.csv</span><span>'</span><span>,</span> <span>index</span><span>=</span><span>False</span><span>)</span>
<span>import</span> <span>pandas</span> <span>as</span> <span>pd</span>
<span>from</span> <span>sklearn.preprocessing</span> <span>import</span> <span>MinMaxScaler</span>

<span># Load data </span><span>data</span> <span>=</span> <span>pd</span><span>.</span><span>read_csv</span><span>(</span><span>'</span><span>historical_data.csv</span><span>'</span><span>)</span>

<span># Handle missing values </span><span>data</span> <span>=</span> <span>data</span><span>.</span><span>dropna</span><span>()</span>

<span># Normalize data </span><span>scaler</span> <span>=</span> <span>MinMaxScaler</span><span>()</span>
<span>data</span><span>[[</span><span>'</span><span>Open</span><span>'</span><span>,</span> <span>'</span><span>High</span><span>'</span><span>,</span> <span>'</span><span>Low</span><span>'</span><span>,</span> <span>'</span><span>Close</span><span>'</span><span>,</span> <span>'</span><span>Volume</span><span>'</span><span>]]</span> <span>=</span> <span>scaler</span><span>.</span><span>fit_transform</span><span>(</span><span>data</span><span>[[</span><span>'</span><span>Open</span><span>'</span><span>,</span> <span>'</span><span>High</span><span>'</span><span>,</span> <span>'</span><span>Low</span><span>'</span><span>,</span> <span>'</span><span>Close</span><span>'</span><span>,</span> <span>'</span><span>Volume</span><span>'</span><span>]])</span>

<span># Save preprocessed data </span><span>data</span><span>.</span><span>to_csv</span><span>(</span><span>'</span><span>preprocessed_data.csv</span><span>'</span><span>,</span> <span>index</span><span>=</span><span>False</span><span>)</span>
import pandas as pd
from sklearn.preprocessing import MinMaxScaler

# Load data data = pd.read_csv('historical_data.csv')

# Handle missing values data = data.dropna()

# Normalize data scaler = MinMaxScaler()
data[['Open', 'High', 'Low', 'Close', 'Volume']] = scaler.fit_transform(data[['Open', 'High', 'Low', 'Close', 'Volume']])

# Save preprocessed data data.to_csv('preprocessed_data.csv', index=False)

Enter fullscreen mode Exit fullscreen mode

Model Building

1. Define the Model: Choose a machine learning algorithm suitable for time series prediction. Common choices include LSTM (Long Short-Term Memory) and GRU (Gated Recurrent Unit) networks.

Example model definition:

<span>import</span> <span>tensorflow</span> <span>as</span> <span>tf</span>
<span>from</span> <span>tensorflow.keras.models</span> <span>import</span> <span>Sequential</span>
<span>from</span> <span>tensorflow.keras.layers</span> <span>import</span> <span>LSTM</span><span>,</span> <span>Dense</span><span>,</span> <span>Dropout</span>
<span>model</span> <span>=</span> <span>Sequential</span><span>()</span>
<span>model</span><span>.</span><span>add</span><span>(</span><span>LSTM</span><span>(</span><span>units</span><span>=</span><span>50</span><span>,</span> <span>return_sequences</span><span>=</span><span>True</span><span>,</span> <span>input_shape</span><span>=</span><span>(</span><span>X_train</span><span>.</span><span>shape</span><span>[</span><span>1</span><span>],</span> <span>1</span><span>)))</span>
<span>model</span><span>.</span><span>add</span><span>(</span><span>Dropout</span><span>(</span><span>0.2</span><span>))</span>
<span>model</span><span>.</span><span>add</span><span>(</span><span>LSTM</span><span>(</span><span>units</span><span>=</span><span>50</span><span>,</span> <span>return_sequences</span><span>=</span><span>False</span><span>))</span>
<span>model</span><span>.</span><span>add</span><span>(</span><span>Dropout</span><span>(</span><span>0.2</span><span>))</span>
<span>model</span><span>.</span><span>add</span><span>(</span><span>Dense</span><span>(</span><span>units</span><span>=</span><span>1</span><span>))</span>
<span>model</span><span>.</span><span>compile</span><span>(</span><span>optimizer</span><span>=</span><span>'</span><span>adam</span><span>'</span><span>,</span> <span>loss</span><span>=</span><span>'</span><span>mean_squared_error</span><span>'</span><span>)</span>
<span>import</span> <span>tensorflow</span> <span>as</span> <span>tf</span>
<span>from</span> <span>tensorflow.keras.models</span> <span>import</span> <span>Sequential</span>
<span>from</span> <span>tensorflow.keras.layers</span> <span>import</span> <span>LSTM</span><span>,</span> <span>Dense</span><span>,</span> <span>Dropout</span>

<span>model</span> <span>=</span> <span>Sequential</span><span>()</span>
<span>model</span><span>.</span><span>add</span><span>(</span><span>LSTM</span><span>(</span><span>units</span><span>=</span><span>50</span><span>,</span> <span>return_sequences</span><span>=</span><span>True</span><span>,</span> <span>input_shape</span><span>=</span><span>(</span><span>X_train</span><span>.</span><span>shape</span><span>[</span><span>1</span><span>],</span> <span>1</span><span>)))</span>
<span>model</span><span>.</span><span>add</span><span>(</span><span>Dropout</span><span>(</span><span>0.2</span><span>))</span>
<span>model</span><span>.</span><span>add</span><span>(</span><span>LSTM</span><span>(</span><span>units</span><span>=</span><span>50</span><span>,</span> <span>return_sequences</span><span>=</span><span>False</span><span>))</span>
<span>model</span><span>.</span><span>add</span><span>(</span><span>Dropout</span><span>(</span><span>0.2</span><span>))</span>
<span>model</span><span>.</span><span>add</span><span>(</span><span>Dense</span><span>(</span><span>units</span><span>=</span><span>1</span><span>))</span>

<span>model</span><span>.</span><span>compile</span><span>(</span><span>optimizer</span><span>=</span><span>'</span><span>adam</span><span>'</span><span>,</span> <span>loss</span><span>=</span><span>'</span><span>mean_squared_error</span><span>'</span><span>)</span>
import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import LSTM, Dense, Dropout

model = Sequential()
model.add(LSTM(units=50, return_sequences=True, input_shape=(X_train.shape[1], 1)))
model.add(Dropout(0.2))
model.add(LSTM(units=50, return_sequences=False))
model.add(Dropout(0.2))
model.add(Dense(units=1))

model.compile(optimizer='adam', loss='mean_squared_error')

Enter fullscreen mode Exit fullscreen mode

Training the Model

1. Split the Data: Split the data into training and testing sets.

<span>from</span> <span>sklearn.model_selection</span> <span>import</span> <span>train_test_split</span>
<span>X</span> <span>=</span> <span>data</span><span>[[</span><span>'</span><span>Open</span><span>'</span><span>,</span> <span>'</span><span>High</span><span>'</span><span>,</span> <span>'</span><span>Low</span><span>'</span><span>,</span> <span>'</span><span>Close</span><span>'</span><span>,</span> <span>'</span><span>Volume</span><span>'</span><span>]].</span><span>values</span>
<span>y</span> <span>=</span> <span>data</span><span>[</span><span>'</span><span>Close</span><span>'</span><span>].</span><span>values</span>
<span>X_train</span><span>,</span> <span>X_test</span><span>,</span> <span>y_train</span><span>,</span> <span>y_test</span> <span>=</span> <span>train_test_split</span><span>(</span><span>X</span><span>,</span> <span>y</span><span>,</span> <span>test_size</span><span>=</span><span>0.2</span><span>,</span> <span>random_state</span><span>=</span><span>42</span><span>)</span>
<span>from</span> <span>sklearn.model_selection</span> <span>import</span> <span>train_test_split</span>

<span>X</span> <span>=</span> <span>data</span><span>[[</span><span>'</span><span>Open</span><span>'</span><span>,</span> <span>'</span><span>High</span><span>'</span><span>,</span> <span>'</span><span>Low</span><span>'</span><span>,</span> <span>'</span><span>Close</span><span>'</span><span>,</span> <span>'</span><span>Volume</span><span>'</span><span>]].</span><span>values</span>
<span>y</span> <span>=</span> <span>data</span><span>[</span><span>'</span><span>Close</span><span>'</span><span>].</span><span>values</span>

<span>X_train</span><span>,</span> <span>X_test</span><span>,</span> <span>y_train</span><span>,</span> <span>y_test</span> <span>=</span> <span>train_test_split</span><span>(</span><span>X</span><span>,</span> <span>y</span><span>,</span> <span>test_size</span><span>=</span><span>0.2</span><span>,</span> <span>random_state</span><span>=</span><span>42</span><span>)</span>
from sklearn.model_selection import train_test_split

X = data[['Open', 'High', 'Low', 'Close', 'Volume']].values
y = data['Close'].values

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

Enter fullscreen mode Exit fullscreen mode

1. Train the Model: Fit the model to the training data.

<span>model</span><span>.</span><span>fit</span><span>(</span><span>X_train</span><span>,</span> <span>y_train</span><span>,</span> <span>epochs</span><span>=</span><span>50</span><span>,</span> <span>batch_size</span><span>=</span><span>32</span><span>)</span>
<span>model</span><span>.</span><span>fit</span><span>(</span><span>X_train</span><span>,</span> <span>y_train</span><span>,</span> <span>epochs</span><span>=</span><span>50</span><span>,</span> <span>batch_size</span><span>=</span><span>32</span><span>)</span>
model.fit(X_train, y_train, epochs=50, batch_size=32)

Enter fullscreen mode Exit fullscreen mode

Evaluating the Model

1. Evaluate Performance: Use appropriate metrics to evaluate the model’s performance on the test data.

<span>from</span> <span>sklearn.metrics</span> <span>import</span> <span>mean_squared_error</span>
<span>predictions</span> <span>=</span> <span>model</span><span>.</span><span>predict</span><span>(</span><span>X_test</span><span>)</span>
<span>mse</span> <span>=</span> <span>mean_squared_error</span><span>(</span><span>y_test</span><span>,</span> <span>predictions</span><span>)</span>
<span>print</span><span>(</span><span>f</span><span>'</span><span>Mean Squared Error: </span><span>{</span><span>mse</span><span>}</span><span>'</span><span>)</span>
<span>from</span> <span>sklearn.metrics</span> <span>import</span> <span>mean_squared_error</span>

<span>predictions</span> <span>=</span> <span>model</span><span>.</span><span>predict</span><span>(</span><span>X_test</span><span>)</span>
<span>mse</span> <span>=</span> <span>mean_squared_error</span><span>(</span><span>y_test</span><span>,</span> <span>predictions</span><span>)</span>
<span>print</span><span>(</span><span>f</span><span>'</span><span>Mean Squared Error: </span><span>{</span><span>mse</span><span>}</span><span>'</span><span>)</span>
from sklearn.metrics import mean_squared_error

predictions = model.predict(X_test)
mse = mean_squared_error(y_test, predictions)
print(f'Mean Squared Error: {mse}')

Enter fullscreen mode Exit fullscreen mode

Making Predictions

1. Make Predictions: Use the trained model to make predictions on new data.

<span>new_data</span> <span>=</span> <span>pd</span><span>.</span><span>read_csv</span><span>(</span><span>'</span><span>new_data.csv</span><span>'</span><span>)</span>
<span>new_data_scaled</span> <span>=</span> <span>scaler</span><span>.</span><span>transform</span><span>(</span><span>new_data</span><span>)</span>
<span>predictions</span> <span>=</span> <span>model</span><span>.</span><span>predict</span><span>(</span><span>new_data_scaled</span><span>)</span>
<span>print</span><span>(</span><span>predictions</span><span>)</span>
<span>new_data</span> <span>=</span> <span>pd</span><span>.</span><span>read_csv</span><span>(</span><span>'</span><span>new_data.csv</span><span>'</span><span>)</span>
<span>new_data_scaled</span> <span>=</span> <span>scaler</span><span>.</span><span>transform</span><span>(</span><span>new_data</span><span>)</span>
<span>predictions</span> <span>=</span> <span>model</span><span>.</span><span>predict</span><span>(</span><span>new_data_scaled</span><span>)</span>
<span>print</span><span>(</span><span>predictions</span><span>)</span>
new_data = pd.read_csv('new_data.csv')
new_data_scaled = scaler.transform(new_data)
predictions = model.predict(new_data_scaled)
print(predictions)

Enter fullscreen mode Exit fullscreen mode

Conclusion

This project demonstrates how to build and evaluate an AI model for trading. By following the steps outlined in this README, you can create your own model to analyze and predict trading data.

原文链接：AI Trading Model