DirectLiNGAM¶

Model¶

DirectLiNGAM [1] is a direct method for learning the basic LiNGAM model [2]. It uses an entropy-based measure [3] to evaluate independence between error variables. The basic LiNGAM model makes the following assumptions.

Linearity
Non-Gaussian continuous error variables (except at most one)
Acyclicity
No hidden common causes

Denote observed variables by ${x}_{i}$ and error variables by ${e}_{i}$ and coefficients or connection strengths ${b}_{ij}$. Collect them in vectors ${x}$ and ${e}$ and a matrix ${B}$, respectivelly. Due to the acyclicity assumption, the adjacency matrix ${B}$ can be permuted to be strictly lower-triangular by a simultaneous row and column permutation. The error variables ${e}_{i}$ are independent due to the assumption of no hidden common causes.

Then, mathematically, the model for observed variable vector ${x}$ is written as

$$ x = Bx + e. $$

Example applications are found here. For example, [4] uses the basic LiNGAM model to infer causal relations of health indice including LDL, HDL, and γGT.

References

[1] S. Shimizu, T. Inazumi, Y. Sogawa, A. Hyvärinen, Y. Kawahara, T. Washio, P. O. Hoyer and K. Bollen. DirectLiNGAM: A direct method for learning a linear non-Gaussian structural equation model. Journal of Machine Learning Research, 12(Apr): 1225–1248, 2011.

[2] S. Shimizu, P. O. Hoyer, A. Hyvärinen, and A. J. Kerminen. A linear non-gaussian acyclic model for causal discovery. Journal of Machine Learning Research, 7:2003-2030, 2006.

[3] A. Hyvärinen and S. M. Smith. Pairwise likelihood ratios for estimation of non-Gaussian structural eauation models. Journal of Machine Learning Research 14:111-152, 2013.

[4] J. Kotoku, A. Oyama, K. Kitazumi, H. Toki, A. Haga, R. Yamamoto, M. Shinzawa, M. Yamakawa, S. Fukui, K. Yamamoto, T. Moriyama. Causal relations of health indices inferred statistically using the DirectLiNGAM algorithm from big data of Osaka prefecture health checkups. PLoS ONE,15(12): e0243229, 2020.

Import and settings¶

In this example, we need to import numpy, pandas, and graphviz in addition to lingam.

import numpy as np
import pandas as pd
import graphviz
import lingam
from lingam.utils import make_dot

print([np.__version__, pd.__version__, graphviz.__version__, lingam.__version__])

np.set_printoptions(precision=3, suppress=True)
np.random.seed(100)

['1.16.2', '0.24.2', '0.11.1', '1.5.1']

Test data¶

We create test data consisting of 6 variables.

x3 = np.random.uniform(size=1000)
x0 = 3.0*x3 + np.random.uniform(size=1000)
x2 = 6.0*x3 + np.random.uniform(size=1000)
x1 = 3.0*x0 + 2.0*x2 + np.random.uniform(size=1000)
x5 = 4.0*x0 + np.random.uniform(size=1000)
x4 = 8.0*x0 - 1.0*x2 + np.random.uniform(size=1000)
X = pd.DataFrame(np.array([x0, x1, x2, x3, x4, x5]).T ,columns=['x0', 'x1', 'x2', 'x3', 'x4', 'x5'])
X.head()

	x0	x1	x2	x3	x4	x5
0	1.657947	12.090323	3.519873	0.543405	10.182785	7.401408
1	1.217345	7.607388	1.693219	0.278369	8.758949	4.912979
2	2.226804	13.483555	3.201513	0.424518	15.398626	9.098729
3	2.756527	20.654225	6.037873	0.844776	16.795156	11.147294
4	0.319283	3.340782	0.727265	0.004719	2.343100	2.037974

m = np.array([[0.0, 0.0, 0.0, 3.0, 0.0, 0.0],
              [3.0, 0.0, 2.0, 0.0, 0.0, 0.0],
              [0.0, 0.0, 0.0, 6.0, 0.0, 0.0],
              [0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
              [8.0, 0.0,-1.0, 0.0, 0.0, 0.0],
              [4.0, 0.0, 0.0, 0.0, 0.0, 0.0]])

dot = make_dot(m)

# Save pdf
dot.render('dag')

# Save png
dot.format = 'png'
dot.render('dag')

dot

Causal Discovery¶

Then, if we want to run DirectLiNGAM algorithm, we create a DirectLiNGAM object and call the fit() method:

model = lingam.DirectLiNGAM()
model.fit(X)

<lingam.direct_lingam.DirectLiNGAM at 0x1f6afac2fd0>

If you want to use the ICA-LiNGAM algorithm, replace DirectLiNGAM above with ICALiNGAM.

Using the causal_order_ property, we can see the causal ordering as a result of the causal discovery.

model.causal_order_

[3, 0, 2, 1, 4, 5]

Also, using the adjacency_matrix_ property, we can see the adjacency matrix as a result of the causal discovery.

model.adjacency_matrix_

array([[ 0.   ,  0.   ,  0.   ,  2.994,  0.   ,  0.   ],
       [ 2.995,  0.   ,  1.993,  0.   ,  0.   ,  0.   ],
       [ 0.   ,  0.   ,  0.   ,  5.586,  0.   ,  0.   ],
       [ 0.   ,  0.   ,  0.   ,  0.   ,  0.   ,  0.   ],
       [ 7.981,  0.   , -0.996,  0.   ,  0.   ,  0.   ],
       [ 3.795,  0.   ,  0.   ,  0.   ,  0.   ,  0.   ]])

We can draw a causal graph by utility funciton.

make_dot(model.adjacency_matrix_)

Independence between error variables¶

To check if the LiNGAM assumption is broken, we can get p-values of independence between error variables. The value in the i-th row and j-th column of the obtained matrix shows the p-value of the independence of the error variables $e_i$ and $e_j$.

p_values = model.get_error_independence_p_values(X)
print(p_values)

[[0.    0.925 0.443 0.978 0.834 0.   ]
 [0.925 0.    0.133 0.881 0.317 0.214]
 [0.443 0.133 0.    0.    0.64  0.001]
 [0.978 0.881 0.    0.    0.681 0.   ]
 [0.834 0.317 0.64  0.681 0.    0.742]
 [0.    0.214 0.001 0.    0.742 0.   ]]