LightweightMMM 🦇 is a lightweight Bayesian Marketing Mix Modeling (MMM) library that allows users to easily train MMMs and obtain channel attribution information.
To generate predictions, we need to provide de media_data_test with impressions or any other media exposure variable. However, how can I provide this data if, for example, impressions for the upcoming week are unknown?
From the examples I've observed, predictions appear to be used exclusively for historical data. How does this contribute to optimizing for the future?
I'm uncertain whether this question is too basic or if the primary purpose of this package is to help us understand how our strategy could have been improved.
To generate predictions, we need to provide de media_data_test with impressions or any other media exposure variable. However, how can I provide this data if, for example, impressions for the upcoming week are unknown?
From the examples I've observed, predictions appear to be used exclusively for historical data. How does this contribute to optimizing for the future?
I'm uncertain whether this question is too basic or if the primary purpose of this package is to help us understand how our strategy could have been improved.