StatisticsLive Code Editor

76 researchers ran this analysis this month

Mixed-Effects Model Visualization in Python

Technique overview

Visualize repeated-measures and clustered data with mixed-effects models, fixed effects, random intercepts, and confidence bands.

Mixed-effects models handle repeated-measures, longitudinal, and nested data where observations are not independent. Instead of pretending every row is a separate subject, the model separates fixed effects, such as treatment or time, from random effects, such as subject-specific baselines or site-specific shifts. The most useful visualization shows both the raw grouped trajectories and the model-implied population trend, so readers can see why a mixed model was needed.

Key points

Visualize repeated-measures and clustered data with mixed-effects models, fixed effects, random intercepts, and confidence bands.
Mixed-effects models handle repeated-measures, longitudinal, and nested data where observations are not independent.
Instead of pretending every row is a separate subject, the model separates fixed effects, such as treatment or time, from random effects, such as subject-specific baselines or site-specific shifts.
The most useful visualization shows both the raw grouped trajectories and the model-implied population trend, so readers can see why a mixed model was needed.

statsmodelsnumpymatplotlibseaborn

Example Visualization

Open full screen

Review the example first, then use the live editor below to run and customize the full workflow.

Mathematical Foundation

Mixed-effects models handle repeated-measures, longitudinal, and nested data where observations are not independent.

Equation

y_ij = beta0 + beta1*x_ij + b0_i + error_ij

Parameter breakdown

beta0Fixed population intercept

beta1Fixed population slope

b0_iRandom intercept for subject or cluster i

error_ijObservation-level residual

When to use this technique

Use mixed-effects models when measurements are repeated within subjects, nested within batches, clustered by site, or otherwise correlated.

Apply This Technique Now

Run this analysis workflow with AI in seconds. Use the prepared technique prompt or bring your own dataset.

Try Technique Prompt Use Your Own Data

View example prompt

Example AI Prompt

"Visualize my mixed-effects model with fixed-effect trends, random intercepts, and confidence bands for each group, then annotate the key model terms"

How to apply this technique in 30 seconds

Upload Data

Upload your CSV or Excel file in Analyze and keep your column names as-is.

Generate

Run the example prompt and let AI generate this technique automatically.

Refine and Export

Adjust code or prompt, then export publication-ready figures.

Implementation Code

The core data processing logic. Copy this block and replace the sample data with your measurements.

import numpy as np
import pandas as pd
import statsmodels.formula.api as smf

np.random.seed(12)
subjects = np.repeat(np.arange(18), 5)
time = np.tile(np.arange(5), 18)
treatment = np.repeat(np.random.choice(["control", "treated"], 18), 5)
subject_shift = np.repeat(np.random.normal(0, 1.0, 18), 5)
treated_effect = (treatment == "treated") * 1.2
response = 5 + 0.45 * time + treated_effect + subject_shift + np.random.normal(0, 0.55, len(time))

df = pd.DataFrame({"subject": subjects, "time": time, "treatment": treatment, "response": response})
model = smf.mixedlm("response ~ time * treatment", df, groups=df["subject"])
result = model.fit()
print(result.summary())

Visualize This Implementation in Analyze

Visualization Code

Complete matplotlib code for a publication-ready figure. Copy, paste into your notebook, and adjust labels to match your data.

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import statsmodels.formula.api as smf

np.random.seed(12)
subjects = np.repeat(np.arange(18), 5)
time = np.tile(np.arange(5), 18)
treatment = np.repeat(np.random.choice(["control", "treated"], 18), 5)
subject_shift = np.repeat(np.random.normal(0, 1.0, 18), 5)
response = 5 + 0.45 * time + (treatment == "treated") * 1.2 + subject_shift + np.random.normal(0, 0.55, len(time))
df = pd.DataFrame({"subject": subjects, "time": time, "treatment": treatment, "response": response})

result = smf.mixedlm("response ~ time * treatment", df, groups=df["subject"]).fit()
pred_grid = pd.DataFrame({
    "time": np.tile(np.linspace(0, 4, 80), 2),
    "treatment": np.repeat(["control", "treated"], 80),
})
pred_grid["prediction"] = result.predict(pred_grid)

fig, ax = plt.subplots(figsize=(8, 5))
for subject, sub in df.groupby("subject"):
    color = "#9240ff" if sub["treatment"].iloc[0] == "treated" else "#888888"
    ax.plot(sub["time"], sub["response"], color=color, alpha=0.22, lw=1)
for label, sub in pred_grid.groupby("treatment"):
    color = "#9240ff" if label == "treated" else "#111111"
    ax.plot(sub["time"], sub["prediction"], color=color, lw=3, label=f"{label} fixed effect")
ax.set_xlabel("Time")
ax.set_ylabel("Response")
ax.set_title("Mixed-Effects Model: Subject Trajectories and Fixed Trends")
ax.legend(frameon=False)
ax.spines[["top", "right"]].set_visible(False)
plt.tight_layout()
plt.savefig("mixed_effects_model_visualization.png", dpi=300, bbox_inches="tight")
plt.show()

Visualize This Code in Analyze

Random Intercept Caterpillar Plot

Plotting subject-level random intercepts helps identify clusters that sit consistently above or below the population trend.

random_intercepts = {
    subject: effects["Group"]
    for subject, effects in result.random_effects.items()
}
ordered = sorted(random_intercepts.items(), key=lambda item: item[1])
print(ordered[:3], ordered[-3:])

Visualize This Advanced Variant

Common Errors and How to Fix Them

Ignoring repeated measurements

Why: Treating clustered observations as independent underestimates uncertainty.

Fix: Use a random intercept for the subject, site, batch, or cluster that creates dependence.

Only showing the fitted line

Why: Readers cannot see the grouped structure that motivated the model.

Fix: Plot raw subject trajectories behind the population-level fitted trend.

Random effects structure is too complex

Why: Small datasets may not support random slopes and multiple grouping terms.

Fix: Start with a random intercept, then add complexity only when justified by design and diagnostics.

Frequently Asked Questions

Apply Mixed-Effects Model Visualization in Python to Your Data

Upload your dataset and Plotivy generates the Python code, runs the analysis, and produces a publication-ready figure.

Generate Code for This Technique

Python Libraries