Gamma distribution is an exponential family distribution for nonnegative data.
Fig. 5 Target family distribution in the dataset Fig. 6 The physicochemical property distribution.
Also, depth information, on-line auto-regressive modeling, and Gaussian family distribution were used to eliminate background regions[26 28].
Direct costs were estimated using a GLM with a log-link function and Poisson family distribution, as determined by modified Park test [40].
To further evaluate whether SVM identifies Src inhibitors rather than membership of certain compound families, compound family distribution of the identified Src inhibitors and non-inhibitors were analyzed.
For example, if the GLM over-dispersion parameter δ ^ is a non-integer, then choosing the closest family distribution could lead to efficiency losses [16],[16].
If the pdf of the raw-scale residuals from GLM are monotonically declining then the appropriate family distribution should be identified using a modified Park test.
From a combined study of Raman and absorption spectroscopies and transmission electron microscopy, we have developed a semi-quantitative way of estimating the diameter and chiral family distribution of the SWCNTs produced.
We conjecture that this is because the natural (canonical) link function for a Gaussian family distribution, especially with small samples, is an identity one – not to mention the precision losses from GLM in the face of heavy-tailed residuals.
The appropriate family distribution is often identified using the so-called modified Park test, that involves a log-gamma GLM regression of var(ADMINCOST) on ln ADM IN ^ COST.
We found that the raw-scale residuals from a GLM with log-link and gamma family distribution did not exhibit a monotonically declining pdf albeit the residuals were kurtotic (coefficient of kurtosis = 4.3907).
