Time-series forecasting — ARIMA and why most series are not stationary
ARIMA and its relatives forecast a series from its own past, but the whole method rests on stationarity — a property financial and on-chain series almost never have without differencing.
Not yet scoped.
Why
The lesson is diagnostic before predictive: test for stationarity, difference until it holds, and see how a naive forecast on a trending series manufactures confidence it has not earned.
How it works
Not yet built.
Where it lands in Jayverse
Verex: test stationarity before forecasting any price series. Run an ADF test and difference the CLOB price or resolution-source history before fitting anything ARIMA-style, since a naive trend forecast on a non-stationary series manufactures false confidence.
Number: log the stationarity check as part of a reading's methodology. Any Number reading built on ARIMA-style forecasting should record its stationarity test and differencing order alongside the reading, tying into the "reading as a licensed, versioned object" rule.
DeFi: check staking-APY series for stationarity before modeling them. The liquid-staking study build should not fit a predictive model on raw APY history without first testing and differencing it.
Key expressions
Words and phrases from this page worth keeping, with the Korean meaning and the sentence they come from.
Expression
뜻 · 쓰이는 자리
rest on
~에 기반하다, ~에 달려있다 · 방법론 전체가 의존하는 전제를 가리킴. "the whole method rests on stationarity"
differencing
차분(연속값의 차이를 구해 시계열을 안정화하는 기법) · 통계·시계열 용어. "almost never have without differencing"
diagnostic before predictive
예측보다 진단이 먼저다 · 검증 절차가 예측 작업보다 우선한다는 원칙. "The lesson is diagnostic before predictive"
manufacture confidence
근거 없는 확신을 만들어내다(조작하다) · 추세가 있는 시계열이 과신을 유발하는 것을 비판. "manufactures confidence it has not earned"
relatives (그 계열 기법들)
유사 기법들, 같은 계열의 방법 · ARIMA와 비슷한 방법론군을 가리킴. "ARIMA and its relatives"
ARIMA
자기회귀누적이동평균(AutoRegressive Integrated Moving Average) · 시계열을 자신의 과거값으로 예측하는 대표적 통계 모델. "ARIMA and its relatives forecast a series from its own past"
stationarity
정상성 · 시계열의 평균·분산이 시간에 따라 변하지 않는 성질, ARIMA가 전제하는 핵심 조건. "the whole method rests on stationarity"