Your English writing platform
Discover LudwigSuggestions(1)
Exact(2)
The models were fit using software default values (R package mgcv, Wood 2006) for the spline basis dimensions of k = 10 for the 1-dimensional and k = 30 for the 2-dimensional splines.
Parameter k is the maximum allowed choice for the basis dimensions (TPRS in our case) and is equivalent to setting the maximum possible degrees of freedom for each term in a model.
Similar(58)
Based on expert knowledge the flexibility of the 1-dimensional splines was validated as sufficient and the default spline basis dimension of k = 10 was not increased.
Again model improvement was validated by a stepwise integration of predictors, and by increasing the basis dimension of the 2-dimensional spatial trend function (Tables 3 and 4).
Models of transformed internode intervals as a function of time were fitted assuming a basis dimension (k) of 50 knots.
The gam approach required the choice of basis dimension number, which was varied from 3 (less smooth) to 6 (more smooth).
Our choice of basis dimension was as large as practical, to allow the data to determine the complexity of the fitted functions.
Instead, we specified h(t) as a thin-plate spline (basis dimension k = 0.9 T) which represents the adjusted smoothed mean value across all sites for each week.
In these univariate models, we evaluated both linear terms (β z i ) and nonlinear smooth functions (f z i ); using a thin-plate spline basis with basis dimension k = 4) of each GIS covariate z i.
g t, r (s i ) accounts for residual monthly spatial variability within the region, specified as spatial bivariate thin-plate penalized spline terms with basis dimension k t, r = I t, r * 0.9.
The value of 0.9 was chosen such that the basis dimension was as large as practicable which allowed the data to determine the complexity of the fitted functions, but was essentially arbitrary.
Write better and faster with AI suggestions while staying true to your unique style.
Since I tried Ludwig back in 2017, I have been constantly using it in both editing and translation. Ever since, I suggest it to my translators at ProSciEditing.

Justyna Jupowicz-Kozak
CEO of Professional Science Editing for Scientists @ prosciediting.com