Dictionary · Glossary of Statistical Terms

The most widely used measure of dispersion of a frequency distribution introduced by K. Pearson (1893). It is equal to the positive square root of the variance. The standard deviation should not be confused with the root mean square deviation.

Numerical data relating to an aggregate of individuals; the science of collecting, analysing and interpreting such data

The adjective “stochastic” implies the presence of a random variable; e.g. stochastic variation is variation in which at least one of the elements is a variate and a stochastic process is one wherein the system incorporates an element of randomness as opposed to a deterministic system.

Taxonomy refers to classification according to presumed natural relationships among types and their subtypes.

A variable is a characteristic of a unit being observed that may assume more than one of a set of values to which a numerical measure or a category from a classification can be assigned.

The variance is the mean square deviation of the variable around the average value. It reflects the dispersion of the empirical values around its mean.

Closeness of computations or estimates to the exact or true values that the statistics were intended to measure.

Data confidentiality is a property of data, usually resulting from legislative measures, which prevents it from unauthorized disclosure.

The context is the circumstances, purpose, and perspective under which an object is defined or used.

(computational bias) An effect which deprives a statistical result of representativeness by systematically distorting it, as distinct from a random error which may distort on any one occasion but balances out on the average.

Characteristics or information, usually numerical, that are collected through observation.

Data analysis is the process of transforming raw data into usable information, often presented in the form of a published analytical article, in order to add value to the statistical output.

Specification of what may be contained in a data or metadata set in terms of the content or, for data only, in terms of the set of key combinations to which specific attributes (defined by the data structure) may be attached.

In its most general sense correlation denoted the interdependence between quantitative or qualitative data. In this sense it would include the association of dichotomised attributes and the contingency of multiply-classified attributes.

An expert system is an intelligent computer program that uses knowledge and inference procedures to solve problems that are difficult enough to require significant human expertise for their solution.

The dimensions of the IMF definition of "data quality" are: - integrity; - methodological soundness; - accuracy and reliability; - serviceability; - accessibility. There are a number of prerequisites for quality. These comprise: - legal and institutional environment; - resources; - quality awareness.

A term used generally to refer to testing significance when specific alternatives to the null hypothesis are considered.

The difference between the observed value of an index and its “true” value. Errors maybe random or systematic. Random errors are generally referred to as “errors”. Systematic errors are called “biases”.

Fitting is the process of verifying whether the data item value is in the previously specified interval.

A continuous monitoring of the process of compilation and of the results of this process.

Refers to the significance of a matter in relation to a set of financial or performance information. If a matter is material to the set of information, then it is likely to be of significance to a user of that information

A model is a formalised expression of a theory or the causal situation which is regarded as having generated observed data. In statistical analysis the model is generally expressed in symbols, that is to say in a mathematical form, but diagrammatic models are also found. The word has recently become very popular and possibly somewhat over-worked.

The totality of features and characteristics of a product or service that bear on its ability to satisfy stated or implied needs.

Metadata is data that defines and describes other data.

Conceptual procedure in database design that removes redundancy in a complex database by establishing dependencies and relationships between database entities. Normalization reduces storage requirements and avoids database inconsistencies.

An outlier is a data value that lies in the tail of the statistical distribution of a set of data values.

A prototype is an original model constructed to include all the technical characteristics and performances of the new product.

Reliability refers to the closeness of the initial estimated value(s) to the subsequent estimated values.

A “what-if” type of analysis to determine the sensitivity of the outcomes to changes in parameters. If a small change in a parameter results in relatively large changes in the outcomes, the outcomes are said to be sensitive to that parameter.