Team Research Visualization

This project analyzes historical Bitcoin price data from 2019 to 2022 to investigate whether there is a significant correlation between opening and closing prices. The research addresses this question using statistical analysis and visualization techniques on cryptocurrency market data.

The study tests the null hypothesis that Bitcoin opening and closing prices are independent against the alternative hypothesis that a significant correlation exists between them. Using the Spearman correlation method due to non-normal distribution, the research provides insights into market behavior and price predictability.

My contributions focused on data cleaning, which was a critical step to ensure data integrity and analysis reliability. I began by identifying null values in the key columns (Open and Close) and standardized their representation by replacing them with NA. Using the zoo package, I applied the na.approx() function for linear interpolation, filling missing values based on adjacent entries. This approach preserved the dataset continuity while minimizing potential biases.

I re-verified the dataset after imputation to confirm no null values remained and calculated summary statistics for the cleaned data to provide insights into its updated distribution. My work complemented the team efforts in exploratory analysis, visualization, and correlation testing.

• Data loading and preprocessing using R
• Null value identification and standardization to NA format
• Linear interpolation using zoo package to fill missing values
• Verification of data completeness after imputation
• Descriptive statistical analysis of opening and closing prices
• Distribution analysis using histogram visualization
• Exponential decay curve fitting to model price distribution
• Scatter plot analysis to examine Open vs Close relationship
• Spearman correlation testing to measure monotonic relationship

Key findings from the analysis:

• The dataset covers 3 years of historical Bitcoin price data from 2019 to 2022
• The research tested whether there is a significant correlation between Open and Close prices using Spearman correlation
• Histogram visualization revealed the distribution of closing prices
• Exponential decay curve was fitted to understand price spread patterns

Explore the project files and related documents: