Reference: the data was obtained from Cryptocompare API. It's about the Bitcoin price against EUR.
Objectives: to predict if the price will go up or down,
- in the next hour;
- for the next 6 hours, i.e., one prediction per hour;
Workflow (main steps): the analysis was performed using the R to,
- fetch histohour data;
- data preparation;
- model fitting;
- forecasting;
- save workspace.
Used the following R packages: