616-617 / 1983
Brief detour
•
This is a generic problem
•
We know a,b and c are important
•
We know how they relate to each other and outcomes
•
Then we show that x and y are more important than a
•
Now we have x and y, b and c
•
But we don’t know how they relate to each other…
•
Updating multi-parameter models with non-independence of parameters
•
Significant underlying technical challenge