EWMA control chart
Plots an exponentially weighted moving average of the measurements with control limits that widen over the first points; sensitive to small shifts, robust to non-normal data.
Menu: Graphs → More charts → Control chart: EWMA…
When should I use it?
- Monitoring a process for small sustained shifts.
What are the assumptions?
- Values in time order; independent.
How do I run it in StatLab?
- Choose the measurement (and subgroup variable if any).
- Choose the weight λ (0.05–0.25 for small shifts) and the limit width L.
How do I read the output?
| Output | Meaning |
|---|---|
| EWMA line | Weighted average; λ is the weight of the newest value. |
| Limits | Target ± L·σ·√(λ/(2−λ)·(1−(1−λ)^(2i))). |
Worked example
On the practice survey dataset (File → Sample dataset), this command — which the dialog's Paste button also produces — runs the procedure:
PROC vizEwma /y=score /x='' /lambda=0.2 /L=3.
Output tables: Points Outside the EWMA Limits.
Computed by StatLab's validated procedure code when this page was built. The example only illustrates the steps; choose variables that fit your own research question.
What are common mistakes?
- Comparing EWMA points with the raw-value limits of an individuals chart.