The article presents a method to estimate the shape of a distribution curve using a mix of two different techniques: one based on a specific model and another more flexible one. By combining these approaches, the researchers can better understand which model fits the data best. The key idea is to use a parameter that adjusts the balance between the two methods, with the goal of getting close to the true distribution shape. This approach was tested on real data sets and showed promising results, providing insights into when the specific model works well and when it doesn't.