Process Capability Calculator

Process Capability Calculator: run process capability calculations online. Formula, assumptions, and interpretation guide.

A time series calculator analyses data collected at regular time intervals to identify trends (long-term direction), seasonality (recurring patterns), and noise (random variation). It can compute moving averages to smooth data, exponential smoothing for forecasting, and growth rates between periods. Time series analysis is used in financial forecasting, sales planning, economic analysis, weather modelling, and operational capacity planning.

Understanding trend direction and seasonal patterns helps businesses plan inventory, staffing, and budgets more accurately than using simple averages.

Enter your time series values in chronological order (e.g. monthly sales figures).
Select the moving average window (e.g. 3-period or 12-period for monthly data).
Review the smoothed trend line to identify the underlying direction stripped of short-term noise.
Check period-over-period growth rates to quantify how fast values are changing.
For forecasting, apply exponential smoothing with an alpha parameter (0.1–0.3 for stable data, 0.5–0.9 for fast-changing data).

Time series formulas

Simple moving average (SMA): SMAₜ = (xₜ + xₜ₋₁ + ... + xₜ₋ₙ₊₁) / n

Exponential smoothing: Sₜ = α × xₜ + (1−α) × Sₜ₋₁, where α is the smoothing factor (0 < α < 1)

Period growth rate: g = (xₜ / xₜ₋₁) − 1

Compound annual growth rate (CAGR): CAGR = (xₙ / x₀)^(1/n) − 1

Reading time series results

Trend, seasonality, and residuals

If the smoothed trend is rising, values are growing over time; falling means decline. A consistent pattern repeating every 12 months in monthly data indicates seasonality. Values that deviate sharply from the smoothed line are anomalies worth investigating. CAGR converts multi-period growth into a single annual growth rate for easy comparison — a CAGR of 15% means the series grew as if it increased 15% each year on a compounding basis.

Statistics tips and best practices

Use a 12-period moving average for monthly data to eliminate seasonal effects and reveal the pure trend.
Compare year-over-year (YoY) growth rather than month-over-month to eliminate seasonality from growth calculations.
Exponential smoothing gives more weight to recent observations — ideal for data where recent values are more predictive than older ones.
Always plot the raw and smoothed data together to visually validate that your smoothing parameter choice makes sense.

US retail sales exhibit strong seasonality: December typically runs 20–30% above the annual monthly average due to holiday spending.
The S&P 500 has produced a CAGR of approximately 10.5% including dividends over the past 50 years, but individual year returns range from −38% (2008) to +34% (1995).
Web traffic typically shows a weekly seasonality with weekday peaks and weekend troughs for B2B sites — the opposite pattern for consumer entertainment sites.

Common mistakes to avoid

Extrapolating a trend far beyond the data range — even accurate trends break down due to market saturation, competition, or external shocks.
Ignoring seasonality when comparing periods — comparing Q4 revenue to Q1 revenue without seasonal adjustment gives a misleading picture of growth.
Using a moving average window that is too short — a 2-period SMA barely smooths anything; a 12-period SMA reveals true multi-month trends.

Time series forecasts are estimates based on historical patterns and do not guarantee future outcomes. Economic, market, and operational conditions can change in ways that invalidate historical patterns. Forecasts should not be used as the sole basis for financial commitments.