What is the significance of "removal_effects" in ChannelAttribution R package?

147 Views Asked by At

I am trying to build a Markov model using the ChannelAttribution package available in R.

Sample code to run this model is -

M <- markov_model(Data, var_path, var_conv, var_value=NULL, var_null=NULL,
order=1, nsim=NULL, max_step=NULL, out_more=FALSE, sep=">", seed=NULL)

Source documentation says that out_more = TRUE option gives model's removal_effects. I would like to know the significance of this M$removal_effects.

If these are the weights of the input channels in the Data , then should they sum to 1 ?

I have tried to print the sum of these removal_effects and I get a different value every time I run the model

Source documentation - https://cran.r-project.org/web/packages/ChannelAttribution/ChannelAttribution.pdf