missing_view
Back to L1.5 | Browse all axes | Browse all options
Axis
missing_viewon sub-layerL1_5_D_missing_outlier_audit(layerl1_5).
Sub-layer
L1_5_D_missing_outlier_audit
Axis metadata
Default:
'multi'Sweepable: False
Status: operational
Operational status summary
Operational: 4 option(s)
Future: 0 option(s)
Options
heatmap – operational
Visualisation of missing pattern over time.
L1.5.D missing-data visualisation heatmap.
This option configures the missing_view axis on the L1_5_D_missing_outlier_audit sub-layer of L1.5; output is emitted under manifest.diagnostics/l1_5/L1_5_D_missing_outlier_audit/ alongside the other selected views.
When to use
Detecting block-missingness (e.g. all 1980s missing) vs scattered NaNs that influences the choice of imputation method.
References
macroforecast design Part 4: ‘diagnostic layers default-off; non-blocking; produce JSON + matplotlib views attached to manifest.diagnostics/.’
Related options: per_series_count, longest_gap, multi
Last reviewed 2026-05-05 by macroforecast author.
longest_gap – operational
Maximum consecutive-missing run per series.
L1.5.D missing-data visualisation longest_gap.
This option configures the missing_view axis on the L1_5_D_missing_outlier_audit sub-layer of L1.5; output is emitted under manifest.diagnostics/l1_5/L1_5_D_missing_outlier_audit/ alongside the other selected views.
When to use
Critical for forward-fill imputation – long runs distort the imputed values; values > 12 (monthly) typically rule out forward-fill.
References
macroforecast design Part 4: ‘diagnostic layers default-off; non-blocking; produce JSON + matplotlib views attached to manifest.diagnostics/.’
Related options: per_series_count, heatmap, multi
Last reviewed 2026-05-05 by macroforecast author.
multi – operational
Produce all three missingness views.
L1.5.D missing-data visualisation multi.
This option configures the missing_view axis on the L1_5_D_missing_outlier_audit sub-layer of L1.5; output is emitted under manifest.diagnostics/l1_5/L1_5_D_missing_outlier_audit/ alongside the other selected views.
When to use
Comprehensive missingness audit; recommended default for first-time runs.
References
macroforecast design Part 4: ‘diagnostic layers default-off; non-blocking; produce JSON + matplotlib views attached to manifest.diagnostics/.’
Related options: per_series_count, heatmap, longest_gap
Last reviewed 2026-05-05 by macroforecast author.
per_series_count – operational
Bar chart of NaN count per series.
L1.5.D missing-data visualisation per_series_count.
This option configures the missing_view axis on the L1_5_D_missing_outlier_audit sub-layer of L1.5; output is emitted under manifest.diagnostics/l1_5/L1_5_D_missing_outlier_audit/ alongside the other selected views.
When to use
Quick view of where L2.D imputation will work hardest; outliers in this chart flag candidates for dropping.
References
macroforecast design Part 4: ‘diagnostic layers default-off; non-blocking; produce JSON + matplotlib views attached to manifest.diagnostics/.’
Related options: heatmap, longest_gap, multi
Last reviewed 2026-05-05 by macroforecast author.