Interpretation of values
evgkra opened this issue · comments
Hello, thank you for your package!
I cannot find any information on how to interpret the values that one recieves after causal inference.
Is it correct to intuitively assume that the higher the value, the higher the correlation between variables? Is there a maximum value?
Hi, if you are talking about the values in the adjacency matrix, they are direct causal effects. Higher the value, higher the direct causal effect.
Thank you!
I have one more question, is there any good way of implementing deltas in the algorithm? In AMOS for example, one has to set causal effects from T1 to T2 and delta to T2, to 1. Here I can only think of adding prior knowledge abouth paths but there is no way of setting the value to 1 from what I can see
Currently, such a way of setting coefficients to be some specific values is not implemented.
How would you propose to work with delta variables, or would you say that the algorithm should not be used for such?
Though I'm not quite sure what delta refers to, the coefficients from error terms to the corresponding observed variables are automatically set to 1.
Deltas refer to the difference between a variable measured at time-point 1 and time-point 2
I see. The current package would not be able to do such an analysis.
Thank you very much for getting back to me so quickly!