FourierTransform

Here, we show model performance in predicting the bitcoin price and their hyper-parameters.

Prediction

Prices

DATE	CLOSE
2024-11-11	66,767.75
2024-11-18	60,845.97
2024-11-25	59,216.39
2024-12-02	62,058.89

Charts

Model Performance

For scoring, Root Mean Squared Error was used.

Training data

Mean across all targets: 6,045.71

y_CLOSE_7	y_CLOSE_14	y_CLOSE_21	y_CLOSE_28
889.53	4,192.14	9,410.96	9,690.23

Test data

Mean across all targets: 11,817.05

y_CLOSE_7	y_CLOSE_14	y_CLOSE_21	y_CLOSE_28
5,733.78	12,361.62	16,431.36	12,741.42

Model Setup

Below are choices made for the model setup.

Exogenous Data

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

No parameters were selected.

Feature Generator

(Determined by Grid Search)

No parameters were selected.

Feature Scaler

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

param	description	value
log	Whether to take the log of the data.	True
diff	Whether to difference the data.	True
yeo_johnson	Whether to apply Yeo-Johnson transformation to the data.	False
normalize	Whether to normalize the data.	True

Hyper-parameters

(Determined by Grid Search)

param	description	value
n_freq	The number of frequencies to use. If -1, all frequencies are used.	-1