/google-trends-financial-modeling

Predicting index movements with Google Trends search volume alternative data

Primary LanguageJupyter Notebook

Google Trends Financial Modeling

Ming Fong and Alexander Yang

STAT 198 Quantitative Finance

Fall 2020

Introduction

We implement a trading strategy for an index (NASDAQ) based on moving averages of changes in Google Trend data for certain selected keywords.

The notebook can be found in Presentation Notebook.ipynb

Alternative Data

We will use Google Trends search volume data to make predictions about the movements of an index.

Data is downloaded through the Pytrends module using the Google Trends API.

Financial data is from yfinance

Backtesting is done using the Backtesting.py package

Resources

Paper to implement: https://www.researchgate.net/publication/326503702_Algorithmic_Trading_Systems_Based_on_Google_Trends

https://jackdry.com/predicting-realized-volatility-using-google-trends