exploratory data analysis


Also found in: Acronyms, Wikipedia.

Exploratory Data Analysis

(EDA)

[J.W.Tukey, "Exploratory Data Analysis", 1977, Addisson Wesley].

exploratory data analysis

a form of statistical analysis which begins by exploring data rather than testing clearly formulated prior hypotheses. Exploratory data analysis does as it says: it explores the pattern of the data set under analysis, considering its range, level, outliers, batching it before graphing and transforming it. The MINITAB computer package, for example, contains these techniques in its sub-pro grammes. In either qualitative or quantitative forms, the purpose of exploratory data analysis is to follow parallel procedures in the interrogation of statistics, i.e. generating hypotheses through exploring the data before turning to confirmatory statistics to test those hypotheses.
Mentioned in ?
References in periodicals archive ?
John Tukey proposed exploratory data analysis (EDA) as a "practical philosophy" of data analysis that minimizes the researcher's dependence on prior assumptions and maximizes insight into the phenomenon at hand.
Keywords: Atmospheric phenomenon, Ozone layer Depletion, stratospheric region, exploratory data analysis, coefficient of variations.
Software is important for successful implementation of graphical techniques within exploratory data analysis.
Plans are under way for additional enhancements to BRFSS Maps, including advanced exploratory data analysis tools, data histograms, multivariate mapping capabilities, county-level mapping, and trend maps.
data mining' has seen exploratory data analysis become an
In addition to the above mentioned exploratory data analysis functionality, analytic applications are also beginning to provide monitoring and alerting features, which push relevant information to users through various desktop and mobile devices.
a developer of open-portal building-block software, has announced the release of Table Lens for Web, a user interface that offers exploratory data analysis and expands access regardless of a user's platform.
Although basic concepts of exploratory data analysis are well presented, the interested reader should consult the book of Haeglin et al.
Sanderson and Fisher (1994, 1997) have emphasized the strong philosophical connection between ESDA and Tukey's (1977) Exploratory Data Analysis (EDA), particularly the EDA analyst's initial skepticism of numerical summaries.
The formulation of the models after exploratory data analysis increases the significance of the results and enhances their power from the economist's and the policy-maker's point of view.
Through both hypothesis validation and exploratory data analysis, statistical techniques are of fundamental importance.
Jeffrey Baxter, senior lecturer at the Flinders University of South Australia, Sturt Campus, specifically implements Minitab to introduce the fundamental ideas of Exploratory Data Analysis (EDA) to those statistics students encountering such concepts for the first time.

Full browser ?