Model >

Arima

Here, we present the model's performance in predicting the bitcoin price along with their hyper-parameters. For more details on the model, see here.

Prediction

Prices

DATE CLOSE
2025-01-15 95,843.54
2025-01-22 95,646.04
2025-01-29 95,824.80
2025-02-05 95,626.61

Charts

Model Performance

For scoring, Root Mean Squared Error was used.

Training data

Mean across all targets: 5,689.54
D+7 D+14 D+21 D+28
5,647.20 5,476.57 6,389.01 5,245.38

Test data

Mean across all targets: 3,974.06
D+7 D+14 D+21 D+28
1,284.33 1,898.48 7,091.71 5,621.72

Model Setup

Below are choices made for the model setup.

Exogenous Data

(Determined by Forward Selection and heuristically based on the model's assumptions)

param description value
include_commodities Whether to include commodities. False
include_currencies Whether to include currencies. False
include_indices Whether to include market indices. False

Feature Generator

(Determined by Grid Search)

No parameters were selected.

Feature Scaler

(Determined by heuristically, considering the model's assumptions)

param description value
diff Whether to difference the data. False
log Whether to take the log of the data. True
normalize Whether to normalize the data. False

Hyper-parameters

(Determined by Grid Search)

param description value
order The ARIMA order of the model. (1, 0, 0)
seasonal_order The seasonal ARIMAorder of the model. (1, 0, 1, 14)