Moving Average & Smoothing
Smooth a series of numbers with three classic filters. Simple moving average (SMA) averages each window of w consecutive points. Weighted moving average (WMA) weights the most recent point highest (linear weights 1…w). Exponential moving average (EMA) weights all past points with exponentially decaying weights controlled by a smoothing factor α ∈ (0, 1]. The trailing (right-aligned) convention is used: SMA/WMA output length is n − w + 1; EMA length is n. Runs locally in your browser.
Series
Result
SMA: out[i] = (Σⱼ x[i−j]) / w for j = 0…w−1, giving n − w + 1 values — the unweighted mean of the trailing window. WMA: out[i] = Σⱼ (w−j)·x[i−j] / (w(w+1)/2) — the most recent point gets weight w, the oldest weight 1, so recent values dominate. EMA: ema[0] = x[0], ema[i] = α·x[i] + (1−α)·ema[i−1]; α = 1 returns the raw series, smaller α smooths more. Series parsed as numbers separated by commas, spaces, or newlines. Pairs with the N-Gram Frequency and Statistics tools. Everything runs locally — nothing leaves your browser.