Logistic Regression

Explanation & Rationale

The Logistic Regression strategy applies a statistical model to predict whether the stock price will go up or down based on the past 30 days of data. By learning from historical trends, the model assigns probabilities to future price movements and generates buy or sell signals accordingly. This approach is useful for capturing simple linear relationships in market behavior and provides a transparent, interpretable framework for making trading decisions.

Code

'''
Logistic Regression Strategy.

Trains a Logistic Regression model using the past 30 days of data to predict the next day's price movement (up/down).
Learn more @ docs.ubacktest.com/examples/machine-learning/logreg
'''

import pandas as pd
from sklearn.linear_model import LogisticRegression
import numpy as np

def logistic_regression(data, window=30):

    signals = np.zeros(len(data))  # Initialize signals array

    # Create the target variable: 1 if price goes up the next day, -1 if it goes down
    data['target'] = (data['close'].shift(-1) > data['close']).astype(int) * 2 - 1

    # Iterate over the data starting from the window index
    for i in range(window, len(data)):
        # Prepare the features (X) and target (y) for the logistic regression model
        X = np.array(range(i-window, i)).reshape(-1, 1)  # Time index for the last window days
        y = data['target'][i-window:i].values  # Target is whether the next day's price goes up (1) or down (-1)

        # Fit the Logistic Regression model
        model = LogisticRegression()
        model.fit(X, y)

        # Predict the next day's movement (up/down)
        prediction = model.predict(np.array([[i]]))  # Predict the next point (i.e., the 6th day)

        # Signal generation based on prediction (uptrend or downtrend)
        signals[i] = prediction[0]  # Buy signal if predicted 1 (up), else sell signal (-1)

    return signals

def strategy(data):

    # Call the logistic_regression function to get the signals
    data['signal'] = logistic_regression(data)

    return data