View source: R/RealEffectTabLong.R
RealEffectTabLong | R Documentation |
These functions are akin to the RealEffectText
function, but rather than generating a
meaningful portrayal of a single predictor's effect on a model's response variable, the RealEffectTab
functions tabulate effects of specified changes for multiple predictors (scaled or unscaled) and multiple
models as a percent change in the response variable. The "Wide" and "Long" suffixes indicate the type of data
the function is meant to be used with.
RealEffectTabLong( Models, Predictors, UnitChanges, ConfInt = 95, Pvals = T, GroupCol, GroupPat = "^[[:alnum:]]+", ScaleSds = rep(NA, length(Predictors)), PredVects = ScaleSds, Data = NULL, Precision = 2 ) RealEffectTabWide( Models, Predictors, UnitChanges, ConfInt = 95, Pvals = T, ScaleSds = rep(NA, length(Predictors)), PredVects = ScaleSds, Data = NULL, Precision = 2 )
Models |
Either a character vector of glmmTMB model object names, or the unquoted name of a single glmmTMB model object. |
Predictors |
A character vector representing the predictors of interest. Appropriate values can be found
by entering |
UnitChanges |
A numeric vector indicating the unscaled change in a predictor for which effects should be returned in the order of the Predictors vector. |
ConfInt |
A number indicating the desired confidence interval as a percent value. |
Pvals |
A logical value indicating whether or not to append predictor p-values to the output table. |
GroupCol |
A character string indicating the column name containing group membership information. |
GroupPat |
A regular expression capable of extracting group membership information from model names. |
ScaleSds |
A numeric vector indicating the how many standard deviations a predictor has been scaled by in the order of the Predictors vector. If a predictor has not been scaled, its respective ScaleSds value should be NA. |
PredVects |
A character vector specifying the column names of the unscaled predictors. |
Data |
The unquoted name of the table containing the unscaled predictor vectors, and when using
|
Precision |
The number of digits right of the decimal to retain in values shown in the resultant table. Note that this argument only influences the values displayed in the resultant table, and no rounding occurs until all calculations are carried out. |
These functions return a data table with six vectors: the name of the model that
the change in the response variable is associated with, the name of the predictor that the change in the
response variable is associated with, the unscaled change in the predictor, the change in the response
variable expressed as a percentage, and the lower and upper confidence intervals expressed as percentages.
If Pvals = T
, a seventh column with predictor p-values is also included.
data("Epfu_Nb2", "Myev_Nb2", "Epfu_Nb2_Long", "Myev_Nb2_Long", "BatDataWide", "BatDataLong", package = "EcoCountHelper") #RealEffectTabWide Single Model Example RealEffectTabWide(Epfu_Nb2, Predictors = c("YdayScale", "MoonScale"), UnitChanges = c(10, 0.3), ScaleSds = c(2,2), PredVects = c("Yday", "MoonPct"), Data = BatDataWide) #RealEffectTabWide Multiple Model Example RealEffectTabWide(c("Epfu_Nb2", "Myev_Nb2"), Predictors = c("YdayScale", "MoonScale"), UnitChanges = c(10, 0.3), ScaleSds = c(2,2), PredVects = c("Yday", "MoonPct"), Data = BatDataWide) #RealEffectTabLong Single Model Example RealEffectTabLong(Epfu_Nb2_Long, Predictors = c("YdayScale", "MoonScale"), UnitChanges = c(10, 0.3), GroupCol = "Species", GroupPat = "^[[:alpha:]]{4}", ScaleSds = c(2,2), PredVects = c("Yday", "MoonPct"), Data = BatDataLong)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.