spaghetti

Version 1.0.0 | 2026-04-08

spaghetti draws individual-level trajectory plots for long-format repeated-measures data and can layer group means and confidence bands on top. It is built for panels that are too dense for one-line-per-person graph code, so it uses broken-line grouping internally to stay practical on large datasets.

Requirements

• Stata 16 or later
• Long-format data with one row per person-time observation
• A numeric outcome variable and a numeric time variable

Installation

capture ado uninstall spaghetti
net install spaghetti, from("https://raw.githubusercontent.com/tpcopeland/Stata-Tools/main/spaghetti") replace

Commands

Command	Description
spaghetti	Draw longitudinal trajectory plots with optional mean overlays

How It Works

• spaghetti outcome, id() time() draws all individual trajectories in one command.
• Add by() when you want separate groups, or colorby() when you want colors driven by another variable.
• Add mean(bold ci) when you want an overlaid summary trend with optional confidence bands.
• Use sample() to thin very dense panels without changing the mean overlay, which is still computed on the full sample.
• Use highlight(), individual(), refline(), and export() to control emphasis and presentation.

Worked Examples

All examples below use webuse nlswork, so they are runnable immediately after installation.

1. Basic trajectories

webuse nlswork, clear
spaghetti ln_wage, id(idcode) time(year)

2. By-group trajectories with a mean overlay

This is the default presentation many users want: thin background lines for individuals and a bold summary line for each group.

webuse nlswork, clear
spaghetti ln_wage, id(idcode) time(year) by(race) mean(bold ci)

3. Sample a dense panel without changing the mean trend

sample() reduces the number of displayed trajectories, but the mean overlay is still computed before sampling.

webuse nlswork, clear
spaghetti ln_wage, id(idcode) time(year) sample(100) seed(12345) mean(bold)

4. Highlight specific individuals

Use highlight() when a few trajectories deserve emphasis and the rest should fade into the background.

webuse nlswork, clear
spaghetti ln_wage, id(idcode) time(year) ///
    highlight(idcode == 1 | idcode == 2 bgopacity(10))

5. Add a vertical reference line

webuse nlswork, clear
spaghetti ln_wage, id(idcode) time(year) ///
    sample(50) ///
    refline(80, label("Policy change") style(dash))

Notes

• by() supports up to 8 levels.
• colorby() cannot be combined with by() or highlight().
• mean(smooth(lowess)) and mean(smooth(linear)) smooth the mean overlay, not the individual trajectories.
• Standard graph options such as title(), subtitle(), note(), ytitle(), xtitle(), plotregion(), graphregion(), name(), and saving() pass through to the underlying twoway call.

Version History

• 1.0.0 (2026-04-08): Initial Stata-Tools release

Author

Timothy P Copeland, Karolinska Institutet