In spite of the optimism regarding the model fit by CPGLM, this study includes a number of limitations.
However, if we have some a priori knowledge about a parameter's value, we would expect a better model fit by actually constraining it.
We checked the model fit by examining model residuals graphically using binned plots and tested for spatial autocorrelation (package ncf (Bjornstad 2016) based on Moran's I).
The data were collected in linear scale and the percentage of cells in different cell cycle phases was defined by Watson pragmatic model fit by Flow Jo (Tree star Inc).
Given the many possible goodness-of-fit indices that exist, the usual advice is to assess model fit by inspecting several fit indices that derive from different principles (Hox, 2002).
For British Columbia, we find the 'optimal' parameter values are very close to the benchmark values and that we only make a small improvement in model fit by taking advantage of the experimental variation.
We evaluated the degree of model fit by calculating the Pearson correlation coefficient (r) between the observed values and model-predicted values of temporal occurrence.
We assessed model fit by the Akaike Information Criterion AICC).
We assessed model fit by looking at residual variation.
The use of quadratic penalization provides stability to the model fit by overcoming collinearity among variables.
Overall model fit and change in model fit by step were assessed using the R statistic.
