, , , , ,

About the Model

The Self-Weighting Model (SWM) consists in computing local and global weights from the constituent independent variables of a function and, from these, weighted averages for said function. See


Model Advantages

SWM makes possible

1. within-set and between-set comparisons
2. calculation of weighted averages from non-additive quantities
3. acceptance or rejection of candidate weighted averages
4. identification of cases where meta-analysis models and traditional transformations fail

The above is possible by considering variability information (i.e., fluctuations) present in the constituent independent variables of a function. If this information is not available, the model suggests the harmonic mean, a statistic that frequently arises in Science and Engineering, as the candidate weighted average.

Practical Applications

With SWM, weighted averages can be easily computed from non-additive quantities like

1. correlation coefficients
2. coefficients of variations

Other applications for SWM are possible.


To learn more about SWM you may want to read the following tutorials: