Coming from Stata

What's the same

• Command names and abbreviations: su, gen, reg, tab.
• if and in qualifiers, varlists with wildcards and ranges (inc*, q1-q10).
• Comments (*, //, /* */) and line continuation (///).
• Factor variables: i., ib3., ib(last)., c., #, ##, with _b[2.region].
• Local and global macros, foreach, forvalues, while, program define with syntax and marksample.
• Stored results: r(), e(), _b[], _se[], return list, ereturn list.
• Missing values sort above all numbers; extended missing codes .a–.z are kept; return codes work with capture.
• .dta files from version 114 to 119 open directly; save writes version 118 (Stata 14+).

What's better

• Every core, every edition. Expressions, grouping, least squares and the iterative models run in parallel, with results that don't depend on the thread count.
• explain after supported commands walks through the computation with your numbers.
• Graphs take over() and by() everywhere, and violin, coefplot, corrplot are built in. Fonts are embedded in exports.
• Publication tables: esttab is built in, writing Word, LaTeX, Markdown or CSV.
• SPSS files open with import spss, with labels and missing-value codes intact.

What's different

• Graph syntax follows Stata's for the common cases, but not every twoway option is supported yet.
• save writes .dta version 118, which Stata 14 and later open; there's no SPSS writer yet.
• There's no Mata; use matrix commands and programs for now.

What's not there yet

Bootstrap and jackknife, ml, multilevel models, xtlogit/xtpoisson, time series (ARIMA), multiple imputation and SEM are on the roadmap. If one of these blocks you, tell us: it moves things up the list.