Missing-value repair and false trends

Apply imputation strategies by sensor type and compare forward-fill choices.

Apply imputation strategies by sensor type and compare forward-fill choices., your practice guide

Apply imputation strategies by sensor type and compare forward-fill choices.
Predict the reading, then compare it with the measurement.

Python 3 in your browser (JupyterLite)

Python · no install

Apply imputation strategies by sensor type and compare forward-fill choices.

Tier 2 · Web · paste-in setup · No account

Version tested: Python 3.12.7 / Pyodide 0.27.6 in JupyterLite 0.6.4; Chromium real browser capture, 2026-10-09. Date: 2026-10-09.

Open the notebook in your browser and run each Python cell; no install or account is needed.

Three ways to run: use JupyterLite here with no install; run main.py locally from the downloadable lab folder; or open the same notebook in Google Colab.

Open in your browser (new tab)

Steps

Screens captured against Python 3 in your browser (JupyterLite) Python 3.12.7 / Pyodide 0.27.6 in JupyterLite 0.6.4; Chromium real browser capture, 2026-10-09 on 2026-10-09; the tool may have moved on — the text steps are the contract.

  1. Step 1

    Do
    Run the Step 1 notebook cell and inspect the timestamped missing markers.
    You will see
    SYNTHETIC minute-resolution temperature; seed=808; no random draws; minute value_C; 0 20.0; 1 21.0; 2 MISSING
    Why it matters
    An absent record must stay distinguishable from zero.

  2. Step 2

    Do
    Run the Step 2 notebook cell to read the complete-case count and mean.
    You will see
    Naive complete-case summary drops missing rows; retained=5; missing=4; 8:31.0; This mean has no values for outage minutes.
    Why it matters
    Dropping missing rows changes the denominator.

  3. Step 3

    Do
    Run the Step 3 notebook cell to inspect the flagged one-minute forward fill.
    You will see
    Forward-fill the one-minute gap at minute 2; minute value_C provenance; 3 23.0 OBSERVED; minute 2 inherits 21.0 C from minute 1.
    Why it matters
    Short fills remain inferred values and need provenance flags.

  4. Step 4

    Do
    Run the Step 4 notebook cell to inspect the false flat segment during the long gap.
    You will see
    Unbounded forward-fill across the three-minute outage; minute value_C provenance; 7 30.0 OBSERVED; Minutes 4-6 appear flat at 23.0 C without observations.
    Why it matters
    Unbounded fill can make an unobserved interval look stable.

  5. Step 5

    Do
    Run the Step 5 notebook cell to impose a one-minute maximum contiguous gap and compare means.
    You will see
    Maximum contiguous gap=1 minute; reject longer runs; minute value_C provenance; unbounded filled count=9, mean=23.889 C; gap-limited count=6, mean=24.333 C
    Why it matters
    A whole-gap limit avoids partly accepting a long outage.

  6. Step 6

    Do
    Run the Step 6 notebook cell and check which values are observed, filled, or still missing.
    You will see
    RESULT CARD: synthetic temperature; seed=808; original: 5 values, mean=25.000 C; A fill flag is needed to distinguish inference from observation.; This synthetic example does not establish what happened in the outage.
    Why it matters
    The repaired summary needs its missingness rule beside it.

Chapter checks

These questions refer to the chapter’s examples. Use the return links to review their answers.

  1. A PIR motion sensor loses 30 seconds of data during a brief network outage. What is the most appropriate imputation choice?

    Return to the chapter’s knowledge check

  2. A motion sensor (PIR) has 30 seconds of missing data due to a network outage. What is the most appropriate imputation strategy?

    Return to the chapter’s knowledge check

Return to Imputation and Noise Filtering · Browse Labs