Calculate Transfer Entropy
Calculates the transfer entropy from time series X to Y, measuring the reduction in uncertainty about Y’s future when conditioning on X’s past, beyond what Y’s own history provides. Transfer entropy is asymmetric and detects directional information flow, making it ideal for causality detection in financial time series.
Use Cases:
- Detect lead-lag relationships between assets
- Identify directional information flow between markets
- Measure predictive causality in trading signals
- Analyze contagion effects across markets
- Study cross-market dependencies and spillovers
Formula: TE(X→Y) = I(Y_future; X_past | Y_past)
Credits: 5 credits per request (Pro Tier) [Tier: ENTERPRISE, Credits: 10]
Authorizations
API key for authentication. Get your key at https://api.fincept.in/auth/register
Body
Historical time series for variable X (source)
[
0.12,
0.15,
0.11,
0.18,
0.14,
0.16,
0.13,
0.17,
0.15,
0.19
]Historical time series for variable Y (target)
[
0.1,
0.13,
0.09,
0.16,
0.12,
0.14,
0.11,
0.15,
0.13,
0.17
]Time lag for transfer entropy calculation (number of steps)
1
Number of bins for discretization (higher = finer resolution but needs more data)
10
