R/ss_modelfit.R
ss_modelfit_multi.RdWrapper function that runs ss_modelfit() across multiple species. Data is
fit to specified allometric equations for each species, as defined within
ref_table. The full list of allometric equations that may be considered in
ref_table can be found in ?eqns_info and data(eqns_info).
ss_modelfit_multi(
data,
ref_table,
species = "species",
modelcode = "modelcode",
response = "height",
predictor = "diameter"
)Dataframe that contains the variables of interest. Each row is a measurement for an individual tree of a particular species.
Dataframe containing an allometric equation for each tree species, in the form of a model code. Each row is a unique species.
Column name of the species variable in both the dataframes
data and ref_table. Defaults to species.
Column name containing the model codes in ref_table. Refer
to data(eqns_info) for more information on model codes.
Column name of the response variable in data. Defaults to
height.
Column name of the predictor variable in data. Defaults to
diameter.
A list of 2 elements:
List of each species' resulting model object.
Table showing each species' resulting model information.
A dataframe with the following variables:
Name of tree species.
Model code for the allometric equation used.
Parameter estimates.
Geometric mean of the response variable used in calculation of AICc (only for transformed models).
Bias correction factor to use on model predictions (only for transformed models).
Range of the predictor variable within the data used to generate the model.
Range of the response variable within the data used to generate the model.
Residual standard error of the model.
Mean standard error of the model.
Adjusted \(R^2\) of the model.
Sample size (no. of trees used to fit model).
ss_modelfit() to fit a specified model for one species.
ss_modelselect() to select a best-fit model for one species.
ss_modelselect_multi() to select best-fit models across multiple species.
Other single-species model functions:
ss_modelfit(),
ss_modelselect_multi(),
ss_modelselect(),
ss_predict(),
ss_simulate()
# first select best-fit model for all species in data
data(urbantrees)
selected <- ss_modelselect_multi(urbantrees, species = 'species',
response = 'height', predictor = 'diameter')
# use function
results <- ss_modelfit_multi(
urbantrees, # any data with similar species (re-use same data in this case)
ref_table = selected$ss_models_info,
species = 'species', modelcode = 'modelcode',
response = 'height', predictor = 'diameter'
)
results$ss_models[[1]] # model object for first species in list
#>
#> Call:
#> lm(formula = y ~ x)
#>
#> Coefficients:
#> (Intercept) x
#> 6.717 9.464
#>
results$ss_models_info # summary of fitted models
#> species modelcode a b c d
#> 1 Albizia saman lin_w1 6.717431 9.4640964 NA NA
#> 2 Hopea odorata quart_w4 -1.149669 149.3178483 -1522.6122 7908.410
#> 3 Syzygium myrtifolium quart_w2 -4.614100 158.5128488 -748.2492 1638.676
#> 4 Terminalia mantaly expo_w1 1.241983 3.8600410 NA NA
#> 5 Xanthostemon chrysanthus loglog_w1 2.947907 0.5916555 NA NA
#> e response_geom_mean correctn_factor predictor_min predictor_max
#> 1 NA 13.566083 1.000000 0.31194369 1.5278875
#> 2 -14225.442 5.390829 1.000000 0.03183099 0.2928451
#> 3 -1321.221 7.735643 1.000000 0.04138029 0.5665916
#> 4 NA 7.596013 1.000533 0.03501409 0.5602254
#> 5 NA 5.041605 1.000883 0.02864789 0.3533240
#> response_min response_max residual_SE mean_SE adj_R2 n
#> 1 8 20 2.2050 4.7889 0.4276 133
#> 2 2 15 11.9430 2.6339 0.5460 483
#> 3 1 18 2.1674 2.2536 0.6662 353
#> 4 3 18 0.2493 0.0615 0.6702 197
#> 5 2 13 0.2123 0.0449 0.6137 418