Fit and Plot a Sigmoidal Regression Curve — sigmoid

Fits a logistic (sigmoidal) curve using stats::nls() with stats::SSlogis(), optionally excluding selected rows from model fitting while still displaying them on the plot. Included and excluded points can be styled independently, and optional SEM error bars and a 95% confidence interval ribbon are added to the plot.

Usage

sigmoid_regression(
  data,
  x_name,
  y_name,
  sem_name = NULL,
  outlier_indices = NULL,
  point_color = "black",
  point_shape = 16,
  point_size = 3,
  excluded_point_color = "red",
  excluded_point_shape = 16,
  line_color = "blue",
  line_type = "solid",
  line_size = 1,
  ci_fill = "blue",
  text_size = 16,
  x_label = NULL,
  y_label = NULL
)

Arguments

data: A data frame containing all columns used for fitting and plotting.
x_name: A string naming the numeric predictor column in data.
y_name: A string naming the numeric response column in data.
sem_name: Optional string naming the SEM column in data. If provided, y +/- SEM error bars are added.
outlier_indices: Optional integer vector of row indices in data to exclude from model fitting.
point_color: Color for included data points and their error bars.
point_shape: Shape for included data points.
point_size: Size for all points.
excluded_point_color: Color for excluded points and their error bars.
excluded_point_shape: Shape for excluded data points.
line_color: Color of the fitted sigmoidal line.
line_type: Linetype of the fitted line (e.g., "solid", "dashed").
line_size: Line width of the fitted line.
ci_fill: Fill color of the 95% confidence interval ribbon.
text_size: Base text size used in ggplot2::theme_classic().
x_label: Optional custom x-axis label. Defaults to x_name when NULL.
y_label: Optional custom y-axis label. Defaults to y_name when NULL.

Value

A named list with:

model: The fitted nls model object.
coefficients: Named numeric vector of fitted coefficients.
plot: A ggplot object containing points, fit, and confidence band.
predicted_curve: Data frame with x values and fitted/CI values.
fit_data: Data used for model fitting (after exclusions).
excluded_data: Data excluded from fitting and highlighted in the plot.

Examples

my_data <- data.frame(
  logd = c(-2, -1, 0, 1, 2, 3),
  mean = c(0.98, 0.90, 0.60, 0.25, 0.10, 0.05),
  sem = rep(0.02, 6),
)
#> Error in data.frame(logd = c(-2, -1, 0, 1, 2, 3), mean = c(0.98, 0.9,     0.6, 0.25, 0.1, 0.05), sem = rep(0.02, 6), ): argument is missing, with no default

outliers <- c(4, 6)
my_data_result <- sigmoid_regression(
  data = my_data,
  x_name = "logd",
  y_name = "mean",
  sem_name = "sem",
  outlier_indices = outliers
)
#> Error: object 'my_data' not found

my_data_result$coefficients
#> Error: object 'my_data_result' not found
my_data_result$plot
#> Error: object 'my_data_result' not found