Scribd
Upload
36 views4 pages

Bitcoin Price Forecasting Guide

The Bitcoin Price Prediction Project aims to forecast Bitcoin prices using time-series analysis and machine learning techniques. Key requirements include sourcing historical data, feature engineering, implementing traditional and machine learning models, and evaluating results with visualizations. The project must be documented and submitted via a GitHub repository, with a deadline of 7 days from assignment date.

Uploaded by

vermatanisha690
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
36 views4 pages

Bitcoin Price Forecasting Guide

The Bitcoin Price Prediction Project aims to forecast Bitcoin prices using time-series analysis and machine learning techniques. Key requirements include sourcing historical data, feature engineering, implementing traditional and machine learning models, and evaluating results with visualizations. The project must be documented and submitted via a GitHub repository, with a deadline of 7 days from assignment date.

Uploaded by

vermatanisha690
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Bitcoin Price Prediction Project Instructions

Objective: Create a Jupyter Notebook demonstrating Bitcoin price forecasting using time-series analysis and
machine learning techniques.

Core Requirements

1. Data Sourcing & Documentation

o Source historical Bitcoin data independently (minimum 3 years)

o Document data provenance and justify source selection

o Include metadata about:

▪ Price granularity (daily/hourly)

▪ Features collected

▪ Data limitations

2. Feature Engineering

o Derive 5+ technical indicators (e.g., volatility measures, momentum oscillators)

o Demonstrate handling of:

▪ Missing values

▪ Non-stationarity

▪ Temporal dependencies

3. Model Implementation

o Implement two contrasting approaches:

▪ Traditional statistical model (e.g., ARIMA, GARCH)

▪ Machine learning model (e.g., LSTM, Prophet, XGBoost)

o Include hyperparameter tuning process documentation

4. Evaluation & Visualization

o Compare models using:


▪ Time-series cross-validation

▪ Economic signi cance metrics (e.g., hypothetical trading returns)

▪ Statistical error measures (RMSE, MAE)

o Create interactive visualizations showing:

▪ Prediction vs actual plots

▪ Residual analysis

▪ Feature importance

Notebook Structure Guide

# Bitcoin Price Prediction


## 1. Data Overview
- Source description
- Raw data visualization

## 2. Methodology
- Feature engineering rationale
- Model selection justi cation

## 3. Implementation
- Data preprocessing steps
- Model training work ow

## 4. Results
- Comparative performance analysis
- Economic impact assessment

## 5. Conclusion
- Key learnings
- Improvement opportunities

Technical Expectations

✅ Required Elements
fi
fl
fi
• Complete code execution (no empty cells)

• Markdown commentary explaining technical choices

• Version control via Git (daily commits visible)

🚫 Prohibited

• Copy-pasted code without understanding

• Uncited third-party code snippets

• Hardcoded values without explanation

Evaluation Criteria

Category Weight Details

Data Handling 30% Source quality, preprocessing rigor

Model Complexity 25% Appropriate technique selection

Results Validity 25% Statistical & economic signi cance

Documentation Quality 20% Readability, reproducibility

Submission Guidelines

1. GitHub Repository containing:

o Jupyter Notebook (.ipynb)

o [Link]

o Dataset description le

o [Link] with setup instructions

2. Zoom Presentation must demonstrate:

o Live notebook execution

o Interactive prediction visualization


fi
fi
o 5-minute Q&A with technical defense

Deadline: 7 days from project assignment date

You might also like