A residual plot displays the difference between observed and predicted values (residuals) against fitted values or an independent variable in regression analysis. This diagnostic visualization helps identify violations of regression assumptions including non-linearity, heteroscedasticity (non-constant variance), and outliers. A well-fitting model shows residuals randomly scattered around zero with no discernible pattern.
""" anyplot.ai
residual-plot: Residual Plot
Library: plotnine 0.15.4 | Python 3.13.13
Quality: 86/100 | Updated: 2026-05-10
"""
import os
import numpy as np
import pandas as pd
from plotnine import (
aes,
element_line,
element_rect,
element_text,
geom_hline,
geom_point,
geom_smooth,
ggplot,
labs,
scale_color_manual,
theme,
theme_minimal,
)
# Theme tokens (see prompts/default-style-guide.md)
THEME = os.getenv("ANYPLOT_THEME", "light")
PAGE_BG = "#FAF8F1" if THEME == "light" else "#1A1A17"
INK = "#1A1A17" if THEME == "light" else "#F0EFE8"
INK_SOFT = "#4A4A44" if THEME == "light" else "#B8B7B0"
RULE = "rgba(26,26,23,0.10)" if THEME == "light" else "rgba(240,239,232,0.10)"
# Okabe-Ito palette
IMPRINT = [
"#009E73", # Brand green (normal points)
"#C475FD", # Vermillion (outliers)
"#4467A3", # Blue (LOWESS line)
]
# Data - Generate realistic regression data with better labels
np.random.seed(42)
n_points = 150
# House prices regression example
house_size = np.linspace(1000, 5000, n_points) # Square feet
price_true = 150 * house_size + 50000 + np.random.normal(0, 30000, n_points)
# Add a few outliers (priced unusually high or low for their size)
outlier_indices = [20, 75, 130]
price_true[outlier_indices] += np.array([150000, -120000, 100000])
# Fit linear regression using OLS
size_mean = np.mean(house_size)
price_mean = np.mean(price_true)
slope = np.sum((house_size - size_mean) * (price_true - price_mean)) / np.sum((house_size - size_mean) ** 2)
intercept = price_mean - slope * size_mean
price_pred = slope * house_size + intercept
# Calculate residuals
residuals = price_true - price_pred
# Identify outliers (beyond 2 standard deviations)
std_resid = np.std(residuals)
is_outlier = np.abs(residuals) > 2 * std_resid
point_type = np.where(is_outlier, "Outlier", "Normal")
# Create DataFrame
df = pd.DataFrame({"fitted": price_pred, "residuals": residuals, "point_type": point_type})
# Create theme customization
anyplot_theme = theme(
plot_background=element_rect(fill=PAGE_BG, color=PAGE_BG),
panel_background=element_rect(fill=PAGE_BG),
panel_grid_major=element_line(color=INK, size=0.3, alpha=0.10),
panel_grid_minor=element_line(color=INK, size=0.2, alpha=0.05),
panel_border=element_rect(color=INK_SOFT, fill=None),
axis_title=element_text(color=INK, size=20),
axis_text=element_text(color=INK_SOFT, size=16),
axis_line=element_line(color=INK_SOFT),
plot_title=element_text(color=INK, size=24),
legend_background=element_rect(fill=PAGE_BG, color=INK_SOFT),
legend_text=element_text(color=INK_SOFT, size=16),
legend_title=element_text(color=INK, size=18),
figure_size=(16, 9),
)
# Create residual plot
plot = (
ggplot(df, aes(x="fitted", y="residuals", color="point_type"))
+ geom_hline(yintercept=0, color=INK_SOFT, size=1.2, linetype="solid", alpha=0.8)
+ geom_hline(yintercept=2 * std_resid, color=INK_SOFT, size=0.8, linetype="dashed", alpha=0.5)
+ geom_hline(yintercept=-2 * std_resid, color=INK_SOFT, size=0.8, linetype="dashed", alpha=0.5)
+ geom_point(size=4, alpha=0.7)
+ geom_smooth(aes(group=1), method="lowess", color=IMPRINT[2], size=1.5, se=False, span=0.5)
+ scale_color_manual(values={"Normal": IMPRINT[0], "Outlier": IMPRINT[1]}, name="Point Type")
+ labs(x="Fitted Values ($)", y="Residuals ($)", title="residual-plot · plotnine · anyplot.ai")
+ theme_minimal()
+ anyplot_theme
)
# Save
output_path = os.path.join(os.path.dirname(__file__), f"plot-{THEME}.png")
plot.save(output_path, dpi=300)
Part of Residual Plot on anyplot.ai.