.png){kind=icon}
.png){kind=icon}
Return
Visualization of Multivariate Normality: A Cumulant Generating Function Approach
.png){kind=icon}
PRE
AI
DOI：10.1007/s13171-025-00429-2.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
We provide a useful and effective method to visualize and graphically assess the multivariate normality of a dataset. The approach relies on a characterization of multivariate normal distribution, which states that the cumulant generating function is quadratic in its argument if and only if the distribution of the underlying random vector is multivariate normal. Thus, the approach is based on the collective evaluation of third and fourth order derivatives of the cumulant generating function, using which we define a function that can be plotted to visually evaluate the departures from multivariate normality.
Keywords:
Cumulant generating function
Data visualization
Multivariate normality
Journal
S
IF:
0.5
Papers:
29
Citations:
0