Model >

Ets

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	99,384.27
2025-01-22	100,118.68
2025-01-29	101,402.55
2025-02-05	102,151.88

Charts

Model Performance

For scoring, Root Mean Squared Error was used.

Training data

Mean across all targets: 1,125.55

D+7	D+14	D+21	D+28
1,760.52	1,035.23	1,558.91	147.54

Test data

Mean across all targets: 7,133.03

D+7	D+14	D+21	D+28
2,914.52	4,485.23	10,814.51	10,317.88

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.	True
log	Whether to take the log of the data.	True
normalize	Whether to normalize the data.	True

Hyper-parameters

(Determined by Grid Search)

param	description	value
damped_trend	Whether the trend is damped in the ETS model.	False
error	The error type of the ETS model.	add
seasonal	The seasonal component of the ETS model.	add
seasonal_periods	The number of seasonal periods in the ETS model.	14
trend	The trend component of the ETS model.	None