extract_surface function

Extract Surface

Function that transforms functional models from linear or additive functional cox models into afcmSurface or lfcmSurface objects to be plotted.

extract_surface(
  mxFDAobject,
  metric,
  model = NULL,
  r = "r",
  value = "fundiff",
  grid_length = 100,
  analysis_vars,
  p = 0.05,
  filter_cols = NULL
)

Arguments

• mxFDAobject: object of class mxFDA with model model calculated wihtin
• metric: spatial summary function to extract surface for
• model: character string for the name of the model for metric data
• r: Character string, the name of the variable that identifies the function domain (usually a radius for spatial summary functions). Default is "r".
• value: Character string, the name of the variable that identifies the spatial summary function values. Default is "fundiff".
• grid_length: Length of grid on which to evaluate coefficient functions.
• analysis_vars: Other variables used in modeling FCM fit.
• p: numeric p-value used for predicting significant AFCM surface
• filter_cols: a named vector of factors to filter summary functions to in c(Derived_Column = "Level_to_Filter") format

Returns

a 4 element list of either class lfcmSurface or afcmSurface depending on the class of model - Surface: data.frame for term predictions for the surface of the metric * radius area

• Prediction: data.frame for standard error of the terms for the above surface. AFCM models use the p to set the upper and lower standard errors of $\beta_1$

• Metric: character of the spatial summary function used; helps keep track if running many models

• P-value: a numeric value of the input p-value

Examples

#load ovarian mxFDA object
data('ovarian_FDA')

#run the lfcm model
ovarian_FDA = run_fcm(ovarian_FDA, model_name = "fit_lfcm",
                      formula = survival_time ~ age, event = "event",
                      metric = "uni g", r = "r", value = "fundiff",
                      analysis_vars = c("age", "survival_time"),
                      afcm = FALSE)

#extract surface
model_surface = extract_surface(ovarian_FDA, metric = 'uni g',
                                model = 'fit_lfcm',
                                analysis_vars = 'age') #variables in model

Author(s)

Julia Wrobel julia.wrobel@emory.edu

Alex Soupir alex.soupir@moffitt.org