Time Series Smoother

• Enter the displayed example values,or choose the local input files requested by the controls. Optional source files take precedence where labelled.
• Choose Calculate result and inspect the complete result and any reported limitations.
• Use Download result and each companion download link to save complete outputs. Reset clears output and restores example controls.

Ordered observations: [1,2,4,8,16]
Smoother: moving
Moving: window size: 3
Moving: alignment: trailing
Moving: incomplete windows: null
Exponential: alpha: 0.25
Exponential: initial level: first
Exponential: entered prior level: 0

• Return the complete simple moving-average or exponential smoothing series. Moving windows are trailing or odd centered,1–1000 observations,with null or partial boundary windows. Exponential smoothing uses alpha greater than zero and at most1,initialized at the first observation or an entered prior level. Only the selected method fields are used. No missing-value interpolation,forecasting or implicit resampling. Numeric JSON arrays,at most10,000 observations,each within +/-1e12; no strings,nulls or missing values. Download the complete JSON report. All processing stays in your browser. Combined input limit: 400,000 UTF-8 bytes, 100,000 characters per field; report limit: 4 MiB. Floating-point rounding and probability underflow can occur.
• Complete outputs only. Unsupported or oversized input is refused. Follow the displayed job-specific bounds and format policies.