Visualising data distributions with kernel density estimation and reduced chi-squared statistic – Spencer, C.J., Yakymchuk, C., Ghaznavi, M. — (currently under review in Geoscience Frontiers)

The application of frequency distribution statistics to data provides objective means to assess the nature of the data distribution and viability of numerical models that are used to visualize and interpret data. Two commonly used tools are the kernel density estimation and reduced chi-squared statistic used in combination with a weighted mean. Due to the wide applicability of these tools, we produced a Java-based computer application called KDX (pronounced kay-dee-kai) to facilitate the visualization of data and utilizing these numerical tools.

KD𝚾 Application, inset showing supporting file save options (png, jpg, svg, pdf).

Customizable settings of KDX a) Data settings, b) General settings, c) Axis settings

Three examples of data displaying respectively over-dispersion (left), a single population (middle), and underdispersion (right).

A large n example (n = 4000) of detrital zircon U–Pb ages.

You can download KDX from the links below.

macOS (10.3 and later, 26 Mb) (unmounted version here)

Windows (Vista and later, 86 Mb)

The open source code for this application is also available for download at GitHub.

The open access manuscript outlining the details of KDX is currently in press in Geoscience Frontiers.