Getting started

1. Open your data

In the app, choose File → Open. From a script:

use survey.dta, clear              // Stata 114–119
import spss survey.sav, clear       // SPSS, compressed or not
import delimited wages.csv, clear   // CSV and other delimited text

Variables keep their labels, value labels and missing-value codes (.a–.z). Each has a measurement level (nominal, ordinal or scale), shown as a badge in the app.

2. Look before you model

describe
summarize wage educ exper, detail
tabulate educ_cat female, column chi2
histogram wage, kdensity

3. Fit a model

generate lwage = ln(wage)
regress lwage educ c.exper##c.exper i.female i.region, vce(cluster region)
margins female                    // predicted log wage by sex
testparm i.region                 // do regions differ at all?

Every menu item in the app builds a command like these and shows it in the history, so you can learn the syntax by clicking and copy the history as a do-file when you're done.

4. Ask it to explain

Add explain to see how each number was computed from your data:

regress lwage educ exper female, explain

5. Make figures

scatter lwage exper, over(female) fit(qfitci) title("Wages and experience")
graph export figure1.pdf, replace

Every graph takes over() to colour by group and by() for small multiples. Export to SVG, PNG or PDF.

6. Export tables

quietly regress lwage educ
estimates store m1
quietly regress lwage educ exper female, robust
estimates store m2
esttab m1 m2 using table1.docx, se label stats(N r2) replace

The extension picks the format: .docx, .tex, .md or .csv.

7. Save it as a do-file

Put the commands in a .do file and run it with File → Run script,do analysis.do, or from a terminal: modellum analysis.do. Anyone with the free engine can rerun it.

Reference

The full command reference will be published with the source code at launch.Questions? support@modellum.app.