First, a piecewise constant regression function is fitted to the log ratios, using the PCF algorithm described above.
The PCF algorithm provides a useful starting point, as it eliminates (or reduces) through smoothing the random variability owing to the measurement process, while at the same time it fits a piecewise constant regression function to the log ratios which is defined everywhere on the genomic range of the data.
The weights are determined based on the kernel density estimation, which can be regarded as a Nadaraya-Watson estimator, i.e., a kind of local constant regression [7].
For example, the RSAD of fitting a linear quantile regression model must be smaller than or equal to the RSAD of fitting a constant quantile regression model to the same data series.
A second nPLS-DA model was build using only those metabolites which showed relatively constant regression coefficients over time.
Where Y is the predicted response variable (conversion rate, %); Ao, Ai, Aii, Aij are constant regression coefficients of the model, and Xi, Xj (i=1, 3; j=1, 3, i≠j) represent the independent variables (reaction parameters) in the form of coded values.
We have also tried different methods than constant regression, such as local linear regression, quadratic regression to select the CNA regions by the threshold of mean and SD.
Examination of the segmentation produced by piecewise constant regression on Watson -strand of S.cerevisiae yeast chromosome 1 leads to the choice of 152 (average segment size 1500 bp) as a sensible number of breakpoints (Huber et al., 2006).
In the study, we propose a concept of incremental fuzzy models in which fuzzy rules are aimed at compensating discrepancies resulting because of the use of a certain global yet simple model of general nature (such as e.g., a constant or linear regression).
In contrast, if (delta ) (=0), the second term in the right-hand side becomes a constant and the regression boils down to the desire-to-link model".
