Using PCA and Canonical Correlation to Forecast Ethereum Returns

Summary

This paper evaluates dimensionality reduction methods for studying the relationship between Bitcoin and Ethereum and predicting Ethereum returns. The dataset uses log returns and additional covariates. The authors begin with Pearson correlation as a preliminary measure of the link between the two cryptocurrencies, then apply canonical correlation analysis and principal component analysis to reduce the data’s dimensionality and examine cross-asset relationships.

For the predictive task, the study uses Bitcoin features to forecast Ethereum returns and measures the performance of the statistical techniques. The supplied description states the research design but gives no forecast scores, sample dates, model details, or comparison with a baseline. It therefore indicates which methods were assessed, but does not show whether dimensionality reduction improved predictions or whether any relationship would generalize beyond the analyzed data. The focus is statistical linkage and return forecasting between two cryptocurrencies, rather than a complete trading strategy with execution or risk rules.

Key ideas

• The study examines links between Bitcoin and Ethereum using their log returns and added covariates.
• Pearson correlation provides an initial assessment of the relationship between the two assets.
• Canonical correlation analysis and principal component analysis are used for dimensionality reduction.
• The predictive task forecasts Ethereum returns using Bitcoin features.
• The supplied description does not report forecast scores or establish whether the methods improve prediction.

Tags

# Dimensionality reduction for prediction: Application to Bitcoin and Ethereum


# Dimensionality reduction for prediction: Application to Bitcoin and Ethereum









The objective of this paper is to assess the performances of dimensionality reduction techniques to establish a link between cryptocurrencies. We have focused our analysis on the two most traded cryptocurrencies: Bitcoin and Ethereum. To perform our analysis, we took log returns and added some covariates to build our data set. We first introduced the pearson correlation coefficient in order to have a preliminary assessment of the link between Bitcoin and Ethereum. We then reduced the dimension of our data set using canonical correlation analysis and principal component analysis. After performing an analysis of the links between Bitcoin and Ethereum with both statistical techniques, we measured their performance on forecasting Ethereum returns with Bitcoin s features.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.