In this study I have used a single distributional assumption, i.e, the normal distribution, as the underlying distribution for the gene expression values.
When the distributional assumption is severely violated, some distribution-free imputation methods may be considered.
Moreover, this analysis would require that we accept that temporal uncertainty is Gaussian far out in the tails of the distribution where we have little data to support the distributional assumption.
Given the censoring, however, we need to make a distributional assumption on ϵ and we assume that it is distributed N 0,σ).
We will develop the implications of the distributional assumption in detail below.
Similarly, because of the lack of distributional assumption, incorporating multiple inputs or outputs is seamless.
Two prediction algorithms based on a single observed time series and without any distributional assumption are proposed.
In particular, our only distributional assumption is the existence of the group-wise covariance matrices, which may even be singular.
More specifically, if the distributional assumption is valid, a parametric model leads to smaller standard errors of the estimates than would be in the absence of a distributional assumption (Collett 2003).
Although the distributional assumption of the SY method is violated in the category health, both methods yield very similar results.
In contrast, distributional assumption is required for a fully parametric PH model (Kalbfleisch and Prentice 2002; Lawless 2002).
