What is it about?
Nonidentifiability in model calibration and its implications for medical decision making. Nonidentifiability means that there is more than unique solution to the calibration problem. That is, more than one parameter set could make your decision model replicate the target data.
Why is it important?
This is the first article in the medical decision making and health economics literature that formally defines the problem of nonidentifiability. In addition, the paper also provides methods to check for the presence of nonidentifiability and recommendations on potential approaches to get rid or reduce nonidentifiability of present. We test such methods on the calibration of a testbed model of cancer relative survival and a realistic model of the natural history of colorectal cancer
Read the Original
This page is a summary of: Nonidentifiability in Model Calibration and Implications for Medical Decision Making, Medical Decision Making, September 2018, SAGE Publications, DOI: 10.1177/0272989x18792283.
You can read the full text:
The following have contributed to this page