Adrien Alvero
Published: 2022
Total Pages:
Get eBook
In the third chapter, "Fuzzy Bunching", we introduce a new fuzzy bunching approach that is robust to noise. The existing bunching approach identifies the extent of bunching from a sharp spike in the probability density function. In many finance settings, however, the sharp spike could be diffused by data noise. The key idea behind our fuzzy bunching estimator is to identify bunching from the area of a bulge in the cumulative distribution function. The fuzzy bunching approach also avoids density estimation, which makes it easy to implement in sparse data. We provide the theoretical foundation of this approach and illustrate the advantages by using simulated and real data.
