feat(length_warmup): add mcq (#145)

Author: amyheather

pages/guide/output_analysis/length_warmup.qmd

@@ -1125,6 +1125,52 @@ Not relevant - replicating an existing model described in a paper, so just used
 
 ## Test yourself
 
+<div class="h3-tight"></div>
+
+### Quiz
+
+```{r}
+#| echo: false
+library(webexercises) # nolint: library_call_linter
+```
+
+::: {.callout-note}
+
+## What main problem does a warm-up period aim to reduce in steady‑state DES models?
+
+```{r}
+#| output: asis
+#| echo: false
+cat(longmcq(c(
+  answer = "Initialisation bias from starting in an empty or atypical state.",
+  "Numerical rounding errors in random number generators.",
+  "Bias caused by using too many replications."
+)))
+```
+
+:::
+
+::: {.callout-note}
+
+## In the time series inspection approach, why do we plot the cumulative mean rather than raw performance measures over time?
+
+```{r}
+#| output: asis
+#| echo: false
+cat(longmcq(c(
+  "Because raw values are always constant in steady state.",
+  answer = paste0(
+    "Because cumulative means smooth short‑term fluctuations and make it ",
+    "easier to see when metrics stabilise."
+  ),
+  "Because cumulative means completely remove all variability from the data."
+)))
+```
+
+:::
+
+### Activity
+
 Try using/adapting the code above for your own simulation and identify a suitable length of warm-up.
 
 ## References