Closed asha24choudhary closed 1 week ago
This issue is stale because it has been open for 14 days with no activity.
FCI outputs a Partial ancestral graph (PAG) whereas DoWhy expects a standard directed acyclic graph (DAG). You will need to convert your PAG to a simple graph with directed edges. Can you share a visualization of your graph?
This issue is stale because it has been open for 14 days with no activity.
Sure, this is the causal graph that I get.
This is how I generated the data
np.random.seed(11)
alpha = 0.5 beta = -5 delta = 0.4 gamma = -0.7
Z = np.random.normal(loc=1, size=500) T = alpha Z + np.random.normal(loc=2, size=500) M = beta T + np.random.normal(size=500) Y = delta Z + gamma M + np.random.normal(size=500) C = T + Y + np.random.normal(size=500)
df = pd.DataFrame(np.stack((Z, T, M, Y, C), axis=1), columns=['Z', 'T', 'M', 'Y', 'C'])
This issue is stale because it has been open for 14 days with no activity.
This issue was closed because it has been inactive for 7 days since being marked as stale.
Hi. I have a dataset and I do not have ground truth of the data. What I'm doing is causal discovery using FCI from causal learn library for instance and then I get an adjacency matrix like ([[ 0, 0, 0, 2, 0], [ 0, 0, 2, 0, 2], [ 0, 2, 0, 2, -1], [ 1, 0, 1, 0, -1], [ 0, 1, 1, 1, 0]]). then i create a networkx graph save the graph into .gml format (test.gml) and when i do
Create the dowhy Causal Model
Identifying the causal effect
i get NetworkXError: graph should be directed acyclic. what should I do to get rid of this error.
I removed the graph as well and identified the effect, but can I rely on this effect or could you please tell me how shall i proceed. i want to determine the effect of the treatment over the outcome, can i use the graph from causal discovery algorithm or what shall i do?
Thank you in advance :)