Google Trend Enhanced Deep Learning Dataset for Renewable Energy Asset Price Prediction

Overview

This dataset accompanies the research paper titled “A Google Trend Enhanced Deep Learning Model for the Prediction of Renewable Energy Asset Price” by Dr. Nachiketa Mishra, Dr. Lalatendu Mishra, Balaji Dinesh, P M Kavyassree . The study investigates the predictive efficiency of various forecasting models using oil prices and investor sentiment for renewable energy assets, specifically focusing on renewable energy ETFs such as ICLN, PBD, and QCLN.

The dataset contains the processed inputs and raw data used in the analysis, including sentiment indices derived from Google Trends and traditional financial indices.

Citation :

Please cite this dataset as:

Mishra, L., Dinesh, B., Kavyassree, P.M. and Mishra, N., 2024. A Google Trend enhanced deep learning model for the prediction of renewable energy asset price. Knowledge-Based Systems, p.112733.

@bibtex@article{MISHRA2025112733,title = {A Google Trend enhanced deep learning model for the prediction of renewable energy asset price},journal = {Knowledge-Based Systems},volume = {308},pages = {112733},year = {2025},issn = {0950-7051},doi = {https://doi.org/10.1016/j.knosys.2024.112733},url = {https://www.sciencedirect.com/science/article/pii/S0950705124013674},author = {Lalatendu Mishra and Balaji Dinesh and P.M. Kavyassree and Nachiketa Mishra},}

Code : 

Refer Repository URL provided

Directory Structure and Description

📦 data
├── 📂 etf-data
│ ├── 📜 ICLN_INPUT.csv # Input data for ICLN
│ ├── 📜 PBD_INPUT.csv # Input data for PBD
│ ├── 📜 QCLN_INPUT.csv # Input data for QCLN
│ └── 📂 raw-data # Original unprocessed data
│ ├── 📂 market-data # ETF market prices and oil volatility (OVX)
│ ├── 📂 navs # Net Asset Value (NAV) data
│ └── 📂 volatility # Volatility data (GARCH and Moving Average models)
├── 📂 google-trends
│ ├── 📜 keys.txt # Keywords for Google Trends search
│ ├── 📂 trends
│ ├── 📂 first-principal-components # Final Google Trend Index (PCA)
│ ├── 📂 formatted-trends # Cleaned trends data
│ └── 📂 raw-google-trends # Raw fetched Google Trends data

Key Files

ICLN_INPUT.csv, PBD_INPUT.csv, QCLN_INPUT.csv: Processed inputs for the prediction models of each ETF.

raw-data: Contains original data for market prices, NAVs, and volatility measures (GARCH, Moving Average).

google-trends: Data related to Google search trends, including raw, formatted, and the final index derived using Principal Component Analysis (PCA).

Usage Notes

Google Trends Data: The Google Trend Index constructed from the keywords can be found in the first-principal-components folder. This index was a key input in the predictive models and used to construct modified indices in data>*_INPUT.csv’s.

Reproducibility: For reproducing the results from the study, you can directly use the inputs provided under /data to build predictive models.

Modifications: If you aim to modify or extend the dataset, be cautious of the index construction process, particularly around Principal Component Analysis (PCA) in the Google Trends data.

License

This dataset is released under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. You are free to share and adapt the data, provided appropriate credit is given.

Contact Information

For any questions or further information, please contact:

Dr. Nachiketa Mishra: Department of Mathematics, Indian Institute of Information Technology Design and Manufacturing Kancheepuram, India

Dr. Lalatendu Mishra: Department of Management Sciences, Indian Institute of Technology Kanpur, India

Balaji Dinesh: Department of Computer Science, Indian Institute of Information Technology Design and Manufacturing Kancheepuram, India. email : [e-mail]