regressionY Mark

library(mosaicr)

The regressionY mark automatically fits and displays regression lines through data points. It’s invaluable for trend analysis and showing relationships between variables.

Basic Example

# Generate sample data with a clear trend
set.seed(123)
data <- data.frame(
  x = 1:20,
  y = 2 * (1:20) + rnorm(20, 0, 3)
)

spec <- list(
  plot = list(
    list(
      mark = "dot",
      data = list(from = "sample_data"),
      x = "x",
      y = "y",
      fill = "lightblue",
      r = 3
    ),
    list(
      mark = "regressionY",
      data = list(from = "sample_data"),
      x = "x",
      y = "y",
      stroke = "steelblue",
      strokeWidth = 2
    )
  )
)

mosaic(spec, sample_data = data)

Customization Options

Different Regression Types

# Polynomial relationship
set.seed(42)
poly_data <- data.frame(
  x = seq(-3, 3, length.out = 30),
  y = -0.5 * seq(-3, 3, length.out = 30)^2 + 2 * seq(-3, 3, length.out = 30) + rnorm(30, 0, 0.5)
)

spec_poly <- list(
  plot = list(
    list(
      mark = "dot",
      data = list(from = "poly_data"),
      x = "x",
      y = "y",
      fill = "coral",
      r = 3
    ),
    list(
      mark = "regressionY",
      data = list(from = "poly_data"),
      x = "x",
      y = "y",
      stroke = "darkred",
      strokeWidth = 2
    )
  )
)

mosaic(spec_poly, poly_data = poly_data)

Multiple Groups

# Data with different groups
set.seed(789)
grouped_data <- data.frame(
  x = rep(1:15, times = 3),
  y = c(
    2 * (1:15) + rnorm(15, 0, 2),      # Group A: positive slope
    -1.5 * (1:15) + 20 + rnorm(15, 0, 2), # Group B: negative slope
    0.3 * (1:15) + 8 + rnorm(15, 0, 1)    # Group C: slight positive
  ),
  group = rep(c("Group A", "Group B", "Group C"), each = 15)
)

spec_groups <- list(
  plot = list(
    list(
      mark = "dot",
      data = list(from = "grouped_data"),
      x = "x",
      y = "y",
      fill = list(column = "group"),
      r = 3,
      fillOpacity = 0.7
    ),
    list(
      mark = "regressionY",
      data = list(from = "grouped_data"),
      x = "x",
      y = "y",
      stroke = list(column = "group"),
      strokeWidth = 2
    )
  )
)

mosaic(spec_groups, grouped_data = grouped_data)

Using Real Data

# Using penguins data for regression analysis
spec_penguins <- list(
  plot = list(
    list(
      mark = "dot",
      data = list(from = "penguins"),
      x = "body_mass_g",
      y = "flipper_length_mm",
      fill = list(column = "species"),
      r = 3,
      fillOpacity = 0.6
    ),
    list(
      mark = "regressionY",
      data = list(from = "penguins"),
      x = "body_mass_g",
      y = "flipper_length_mm",
      stroke = "black",
      strokeWidth = 2
    )
  ),
  xLabel = "Body Mass (g)",
  yLabel = "Flipper Length (mm)"
)

mosaic(spec_penguins, penguins = penguins)

Use Cases

• Trend Analysis: Identifying and visualizing trends in data
• Predictive Modeling: Showing fitted models alongside data
• Correlation Visualization: Making relationships between variables explicit
• Scientific Research: Standard tool for showing statistical relationships
• Business Intelligence: Forecasting and trend analysis in business data