OLS regression — the line that assumes it caused something
Least-squares regression is the workhorse everyone reaches for first; the danger is that a fitted coefficient reads like a cause when it is only a correlation under stated assumptions.
Not yet scoped.
Why
The point is not the fit but the assumptions behind it — linearity, independent errors, no omitted variable — and the PoC would show the same dataset giving opposite conclusions when one assumption is quietly violated.
How it works
Not yet built.
Where it lands in Jayverse
Number: state the assumptions checked, not just the coefficient. Any reading built on a regression coefficient must publish which assumptions were checked — linearity, error independence, omitted-variable review — in the reading's methodology field, not just the fitted number.
Auditor: flag correlation presented as causal. Add "coefficient without stated assumptions" to the checklist; a published number that skips this fails the check the Auditor row already exists to run.
Key expressions
Words and phrases from this page worth keeping, with the Korean meaning and the sentence they come from.
Expression
뜻 · 쓰이는 자리
workhorse
(가장 흔히 쓰이는) 주력 도구 · 누구나 제일 먼저 손 대는 기본 회귀 기법 · "the workhorse everyone reaches for first"
reach for
(습관적으로) 손을 뻗다, 집어 들다 · 가장 먼저 시도하는 분석 방법이라는 뜻 · "everyone reaches for first"
quietly violated
티 안 나게 어겨지다 · 가정 하나가 몰래 깨졌을 때 결과가 뒤집힘 · "when one assumption is quietly violated"
opposite conclusions
정반대의 결론 · 같은 데이터로도 가정 위반 시 결론이 뒤집힐 수 있음 · "the same dataset giving opposite conclusions"
omitted variable
누락된 변수 · 회귀분석에서 빠뜨리면 인과 해석이 왜곡되는 변수 · "linearity, independent errors, no omitted variable"
OLS
최소제곱법(Ordinary Least Squares, OLS) · 잔차 제곱합을 최소화해 직선을 맞추는 가장 기본적인 회귀 기법, 이 카드의 주제. "OLS regression — the line that assumes it caused something"