Empirical Measurement & Statistical Inference Methods

← Back to Domain-Specific Families

Abstractions that define procedures for measuring, testing, or estimating quantities from data, including statistical significance tests such as the Shapiro-Wilk and D'Agostino's K-squared tests, robust estimators like M-estimators and MAP estimators, and domain-specific diagnostic scales such as Killip class and famine scales.

50 abstractions in this family — domain-specific abstractions that sit near one another in structural-signature space (k-means over structural-signature embeddings). Each is shown with its short description.

• Analytical Method — A repeatable and reviewable procedure that selects inputs, applies explicit transformations or interpretive rules, and produces findings about a defined question under stated assumptions and quality controls.
• Analytical technique — A defined chemical-measurement procedure that prepares a material, produces a selective response to composition, calibrates that response, and identifies or quantifies target components with stated uncertainty and interference limits.
• Appearance event ordination — A quantitative biochronological method that infers a best-fit relative ordering of fossil taxa's first and last appearances from pairwise stratigraphic constraints, then calibrates that event sequence to numerical time.
• Bayesian Programming — A probabilistic-program specification method that defines variables, factorizes their joint distribution, assigns parametric or nested forms, learns unspecified parameters from data, and answers conditional probability queries.
• Bootstrapping populations — A parametric algorithmic-inference method that generates parameter replicas compatible with an observed sample and plugs them into a model family to form candidate populations.
• Bradford Hill criteria — Nine epidemiologic viewpoints used to organize evidence about whether an observed association may be causal, without treating any fixed subset as a necessary, sufficient, or mechanically scored proof of causation.
• Constant false alarm rate — A family of adaptive radar detection algorithms that estimates local noise or clutter from reference cells and scales a detection threshold to maintain a specified false-alarm probability despite changing background power, subject to target-contamination and clutter-model limits.
• Cross-entropy — The expected negative log probability that a model distribution q assigns to outcomes actually drawn from a distribution p.
• Cue Validity — A cue's validity for a category is the chance of category membership among objects bearing that cue in a stated comparison population.
• Cuzick–Edwards Test — A case-control nearest-neighbor significance test for detecting spatial clustering of cases within an already nonuniform background population represented by control locations.
• D'Agostino's K-squared test — An omnibus sample-normality test that transforms sample skewness and kurtosis into approximately standard-normal components and sums their squares, testing an i.i.d. Gaussian null specifically against skewness and tail/peakedness departures.
• Data Mining — The iterative discovery and evaluation of useful, understandable patterns in substantial datasets through coordinated preprocessing, modeling, validation, and deployment.
• Economic Complexity Index — A model-based index inferring the productive capabilities of a location from the diversity and ubiquity pattern of its economic activities, commonly exports.
• Event detection for WSN — A distributed sensing workflow that detects a prespecified environmental or system event at resource-constrained wireless nodes and communicates only qualifying evidence or decisions, trading communication energy against detection delay, misses, false alarms, and network robustness.
• Event Sampling Methodology — A repeated daily-life assessment design that samples participants' current or recent experiences under a declared time-, signal-, or event-contingent protocol.
• False coverage rate — The expected proportion of selected confidence intervals that fail to contain their corresponding true parameters, controlled to address selective reporting in multiple-parameter inference.
• Famine scales — Operational food-security classifications that combine observed severity indicators with explicit thresholds to distinguish adequate conditions, crisis, and famine and to guide response.
• Frequency (statistics) — The absolute count of observations in a declared category or event within a bounded data set; relative frequency divides that count by the total, and cumulative frequency aggregates ordered categories up to a threshold.
• Frequentist Probability — Interpret an event's probability through its stable long-run relative frequency in a specified repeatable reference sequence.
• FWL theorem — An ordinary-least-squares equivalence stating that a regressor's coefficient equals the coefficient obtained after residualizing both the outcome and that regressor against the other included regressors.
• Gaussian Naive Bayes — A naive Bayes classifier for continuous features that models each feature’s class-conditional distribution as Gaussian and combines their likelihoods under conditional independence.
• GC Skew — Measure the signed excess of guanine over cytosine on an oriented sequence segment as their count difference divided by their nonzero count total.
• Gelman–Rubin Statistic — Compare between-chain with within-chain spread to flag incomplete mixing of iterative simulations.
• Genealogical DNA test — A consumer or research genetic comparison that samples selected autosomal, mitochondrial, or Y-chromosome markers to estimate biological relationships, lineage affinities, or reference-population ancestry under probabilistic inheritance and database models.
• Heteroduplex analysis — A mutation-screening method that mixes, denatures, and reanneals reference and test DNA so sequence differences form mismatched heteroduplex molecules whose altered conformation or mobility can be separated from homoduplexes, indicating a variant region without necessarily identifying the exact change.
• Immunoassay — A biochemical analytical procedure that uses specific antibody–antigen recognition to convert the presence or amount of an analyte into an interpretable qualitative or quantitative signal.
• Kaniadakis logistic distribution — A four-parameter continuous distribution on nonnegative values that replaces ordinary exponentials in a generalized logistic form with the κ-exponential, recovering the classical limit as κ approaches zero.
• Killip class — A four-class bedside stratification of acute myocardial infarction by clinical severity of heart failure, from no failure signs through pulmonary edema to cardiogenic shock, historically associated with mortality risk.
• Length time bias — A screening-selection bias in which slowly progressing disease remains detectable for longer and is therefore overrepresented among screen-detected cases, making their observed survival appear better even without a screening benefit.
• M-Estimator — An extremum estimator obtained by optimizing a sample-average criterion—or more generally solving an estimating equation—encompassing maximum likelihood, nonlinear least squares, and many but not inherently robust procedures.
• MAP estimator — A Bayesian point estimator that selects the parameter value maximizing posterior density, combining the data likelihood with a prior and reducing to maximum likelihood only when the prior is constant over the relevant domain.
• Measurement Scale — A rule-governed mapping from empirical instances or states of an attribute into labels, ordered categories, or numerical values whose interpretation preserves declared empirical relations.
• Median Absolute Deviation — A robust measure of univariate dispersion defined as the median of the absolute distances from the sample median, resistant to a minority of extreme observations.
• Microsegment — A very small, behaviorally or contextually defined customer group selected for differentiated prediction, offers, messaging, or treatment at near-individual granularity.
• Mill's Methods — Five comparative methods for inductively identifying possible causal conditions through agreement, difference, combined comparison, residues, and concomitant variation.
• Plant Cover — The proportion of a defined ground plot vertically occupied by the projection of a plant species, vegetation layer, or total vegetation, estimated visually or by point interception under a stated sampling protocol.
• Pseudoreplication — An inferential error that treats nonindependent observations or subsamples as independent experimental replicates, misidentifying the unit of analysis and usually understating uncertainty or confounding treatment with unit effects.
• Rademacher complexity — A sample-dependent measure of how strongly a function class can correlate with independent random ±1 labels, used to bound generalization error.
• Rank product — A nonparametric statistic that combines an item's within-replicate ranks by their geometric mean, often with permutation-based significance estimation to detect consistently high or low differential expression across experiments.
• Rapid Shallow Breathing Index — The ratio of respiratory frequency in breaths per minute to tidal volume in liters, used as one high-level indicator during assessment of readiness to discontinue mechanical ventilatory support, with protocol-dependent thresholds and limited standalone predictive value.
• Representational drift — The gradual change in population-level neural activity patterns encoding stable information or behavior, despite continued functional performance.
• Response-rate ratio — The ratio of the response proportion in a treatment or exposed group to the response proportion in a control or reference group, used as a relative efficacy measure for a prespecified binary outcome over a declared follow-up period.
• Risk Score — A rule-governed numerical or ordinal summary that maps declared predictors to an estimate or stratum of a specified adverse outcome for a stated population, horizon, and use.
• Semiorder — A partial order representable by assigning real-valued utilities and a positive discrimination threshold so that one item is preferred to another only when their scores differ by at least that threshold, allowing nontransitive incomparability.
• Shapiro–Wilk Test — A statistical normality test whose statistic measures how closely ordered sample values align with expected order statistics from a normal population.
• Spectrum Bias — Identify diagnostic-accuracy error when an estimate from one patient spectrum is applied to a different intended spectrum whose conditional test performance differs.
• Standardized Rate — A population event rate adjusted to a declared reference composition or reference rate schedule, so a known difference in population mix does not masquerade as an event-rate difference.
• Stationary Subspace Analysis — A blind-source-separation method that finds linear projections of a multivariate time series whose distributional statistics remain stable across epochs and complementary projections that capture nonstationary change.
• Statistical regularity — The long-run stabilization of frequencies or distributional summaries across many repetitions or sufficiently comparable random events.
• Uncertainty analysis — The systematic identification, quantification, propagation, and communication of uncertainty in measurements, model inputs, assumptions, and outputs used for inference or decisions.