The analysis dashboard

The page

The navbar groups the tabs in menus: Data (Describe, Distributions), Over time (Year over year, Rates, Trends), Causal and survey (Survey, Policy), Capsules (Drift, Capsule, Proof ledger) and Domains (First Nations, Métis and Inuit; Courts and corrections; Air pollution; Hawkes), then Report and Settings. At the right, Share this view gives a link that reopens the page with the same tab, data and form, and the switch sets light or dark mode.

Each analysis tab has its form in a sidebar (the data, the analysis and its arguments) and opens with its first analysis already run on a demo dataset. Its results are in cards:

• Value boxes: the headline estimate, its interval, n (with the p-value) and a verdict.
• Plot: the package’s plot for the result, or the estimates with their intervals when the function has no plot. It opens full screen, and downloads as PNG, SVG or the table as CSV.
• Method: the method in its help page’s words, its assumptions, what the numbers mean, and how to read the effect sizes and fit statistics the result carries (standardised differences, Cramér’s V, I-squared, correlations, ratios, the branching ratio, the time-rescaling test, overdispersion, the PSI).
• Diagnostics: the diagnostic plots the method has (QQ, PP and tail plots of a fit, the time-rescaling QQ plot of a Hawkes fit, observed and fitted values with residuals of a count trend, the clearance interval of a case flow, the fingerprint of a drift), the checks and secondary tables the function returns, and the rows by flag.
• Table, with the printed result below it, and the Reproducible R call.

Some views are linked. Choose a column name in the profile on the Describe tab to see its distribution. Point at the year-over-year plot to see the rows of that period. On the Drift tab, drag across the two densities of a column to see the share of each snapshot’s rows in that range.

The resampling analyses (bootstraps, permutations, simulations, the Hawkes fits) run in a forked background process on Linux and macOS, so the page stays live while they work; the Run button shows that it is busy. A refusal appears in the page in the package’s words, with explerr()’s note on where it comes from, and as a notice in the corner.

Every analysis shows the R call it ran, and it runs exactly that call. Copy the call into a script and you get the same result.

Data

Every tab opens on a dataset from rmbldemo(): made-up, seeded data shaped like what the tab reads. They are the same on every call.

rmbldemo()
#>               name rows
#> 1            court  576
#> 2         offences 1440
#> 3  offences_annual  120
#> 4           survey 1200
#> 5             fnmi  128
#> 6             otis  144
#> 7              air 2016
#> 8           hawkes   NA
#> 9            drift 3000
#> 10           waits 1000
#> 11          ledger  120
#>                                                                  description
#> 1           Court caseload panel, 16 courts by 36 months, a staggered reform
#> 2   Monthly offence counts by area, seasonal, with a step down in April 2020
#> 3                       The offence counts by area and year, with population
#> 4    Stratified two-stage survey with design and bootstrap replicate weights
#> 5  Published health table by distinction, suppressed cells and quality flags
#> 6       Custody admissions and days by institution and year (stock and flow)
#> 7                   Hourly NO2 at six monitors around a stack, with the wind
#> 8                                   Self-exciting event stream over one year
#> 9              Two snapshots of one table, reference and current, with drift
#> 10                  Days to disposition and appearances of 1,000 court cases
#> 11                 Proof ledger of 120 theorems and the functions they cover

The court panel, for example, has a reform adopted by eight of sixteen courts at four different months. It lowers the median delay by 12 days, and the courts share a trend, so the parallel-trends assumption holds:

court <- rmbldemo("court")
plndid(court, unit = "court", time = "month", y = "delay", trtd = "reformed")
#> Difference in differences: 576 observations, 16 clusters
#>     term estimate     se  lower  upper     t   p_value
#>  treated   -12.61 0.4724 -13.62 -11.61 -26.7 4.643e-14
#> Effect relative to the comparison units' pre-period mean (100): -12.6%

The Data sidebar (the arrow at the left) takes a demo dataset, a CSV file (plain or gzip), a capsule directory, or a table from data.rmorie.com, and Load all demos loads every demo dataset so you can switch between them. Data chosen there is used by every tab (“The dataset chosen at the left” in a tab’s Data menu). For a capsule it loads the pinned data file. For data.rmorie.com, sign in first with rmbl login. The app never holds your data key: dload() reads it from where rmbl login stored it.

The tabs

• Describe: profcols(), freqtbl(), cortbl(), mahaoutl() and missgrid(), plus ridgelines, distribution functions, violins, hexagonal bins and concentration curves.
• Distributions: dstfit() fits candidate distributions and ranks them, with density, QQ, PP and tail plots. You can draw a synthetic sample from the best fit with mksyncol().
• Year over year: yoy() with the period, value, groups, the smallest base for a percentage and which direction is better. Download the result as the yoyhtml() page.
• Rates: rtcalc(), ratechg() and shrs(). A rate per population and a share of a total stay apart.
• Trends: trndtst() (Mann-Kendall with the Theil-Sen slope), stepchg(), cnttrnd(), a calendar heatmap, a horizon chart, small multiples and a bump chart.
• Survey: svydes() with weights, strata, clusters or replicate weights, then svyest() or svycmp().
• Policy: the panel designs on unit-by-period data, from plndid() and evstudy() to syndid(), wcb(), plnpwr() and tse(). The tab names the assumption they share: without the policy, the outcomes would have moved in parallel. Read the pre-period coefficients and the cohort trends before the estimate.
• Drift: cpsdrft() on two snapshots, and the app of drftapp() for two CSV files.
• Capsule: a capsule’s provenance, manifest, attestation and bundle files (a demo capsule is built in a temporary directory). Its buttons run vrfcps() and cpschk(), or cpsrfrsh().
• Proof ledger: a ledger from the demo, a file, a capsule or an installed package, its theorems and ldgrchk()’s rows.
• First Nations, Métis and Inuit: Statistics Canada’s tables as indiex() ships them, a demo table in the same layout, or your own. Each distinction is reported on its own. Suppressed cells stay suppressed, and the quality flags are shown. The tab states the OCAP® and CARE principles and links the source tables.
• Courts and corrections: caseflow(), stkflow(), clrdcmp(), dispmix(), lifetbl(), crmdens(), knoxtst(), rgnmapintg(), and the SIU readers siuprs() and siuso() on pasted text.
• Air pollution: a plume map (plmfld()), the demo monitors’ readings, a downwind profile and receptor concentrations (gplume()), the concentration-response functions, polbrdn() and vrfpol().
• Hawkes: chwkfit() with plothwk() (intensity, time-rescaling check, branching tree), chwksim() and chwkjit().
• Report: what you added from the tabs. Save it as one HTML page with the plots as inline SVG, as Markdown, or copy it as a GitHub comment.
• Settings: the theme (light, dark or as the system is set), printed digits, and the seed the seeded analyses use.

The same work in a script

What the Distributions tab runs on its demo data:

wait <- rmbldemo("waits")$days
fit <- dstfit(wait, nboot = 50)
fit$table[, c("rank", "family", "aic", "ks")]
#>   rank      family      aic         ks
#> 1    1       gamma 12144.46 0.04754282
#> 2    2     weibull 12163.66 0.05231905
#> 3    3   lognormal 12166.91 0.04705977
#> 4    4 exponential 12196.68 0.07886116
#> 5    5           t 12571.14 0.13837486
#> 6    6      normal 12934.52 0.16103623
#> 7    7      pareto 14258.49 0.41016644
plot(fit, which = "qq", fams = c("gamma", "weibull"))

And what the Proof ledger tab runs on the sample ledger:

led <- ldgrread(system.file("extdata", "ledger_sample.json", package = "rmoriebricklayer"))
led
#> Proof ledger rmoriebricklayer sample ledger: 3 theorems (1 proved, 1 sorry, 1 axiom)
#>   https://github.com/rootcoder007/rmorie-proofs @ 0123456789ab, Lean v4.12.0
ldgrchk(led)[, c("check", "ok", "severity")]
#>                                                check    ok severity
#> 1                                 proved: gini-equal  TRUE    error
#> 2     covered: gini-equal rmoriebricklayer::ginicoef  TRUE    error
#> 3                                  check: gini-equal  TRUE    error
#> 4                            proved: lorenz-monotone FALSE    error
#> 5 covered: lorenz-monotone rmoriebricklayer::lorenzc  TRUE    error
#> 6                                proved: float-round FALSE     warn

A ledger records the theorems of a formal-proof build and the functions each one covers. Any package can ship one in inst/ledger/, and the tab finds it there. A theorem counts as proved only when no sorry remains. An axiom is assumed, not proved, so it shows as a warning. A ledger pinned in a capsule (provenance kind: "ledger") is checked by rmbl verify too.