causationentropy.datasets package

Synthetic data generation for testing and examples.

causationentropy.datasets.synthetic module

Generate synthetic time series data with known causal structures.

causationentropy.datasets.synthetic.logistic_map(X, r)[source]

causationentropy.datasets.synthetic.logisic_dynamics(n=20, p=0.1, t=100, r=3.99, sigma=0.1, seed=42)[source]: Network coupled logistic map, r is the logistic map parameter and sigma is the coupling strength between oscillators

causationentropy.datasets.synthetic.linear_stochastic_gaussian_process(rho, n=20, T=100, p=0.1, epsilon=0.1, seed=42, G=None)[source]: Linear stochastic Gaussian process

causationentropy.datasets.synthetic.poisson_coupled_oscillators(n=10, T=100, p=0.2, lambda_base=2.0, coupling_strength=0.3, seed=42, G=None)[source]

Coupled Poisson oscillators where each node’s rate depends on its neighbors’ previous states.

Parameters:

n (int) – Number of oscillators
T (int) – Number of time steps
p (float) – Edge probability for random graph
lambda_base (float) – Base Poisson rate
coupling_strength (float) – Strength of coupling between oscillators
seed (int) – Random seed

Returns:

X (array (T, n)) – Time series of Poisson counts
A (array (n, n)) – True adjacency matrix

References

[1] Xanthi Pedeli, Dimitris Karlis, Some properties of multivariate INAR(1) processes, Computational Statistics & Data Analysis. (2013)

Dynamical Systems

causationentropy.datasets.synthetic.logistic_map(X, r)[source]

causationentropy.datasets.synthetic.logisic_dynamics(n=20, p=0.1, t=100, r=3.99, sigma=0.1, seed=42)[source]: Network coupled logistic map, r is the logistic map parameter and sigma is the coupling strength between oscillators

causationentropy.datasets.synthetic.linear_stochastic_gaussian_process(rho, n=20, T=100, p=0.1, epsilon=0.1, seed=42, G=None)[source]: Linear stochastic Gaussian process

Coupled Oscillators