Exploratory data analysis: Difference between revisions

Revision as of 06:12, 3 November 2024

Latest revision as of 20:43, 25 May 2025

(4 intermediate revisions by 4 users not shown)

Line 1:

{{short description|Approach of analyzing data sets in statistics}}{{Data Visualization}}

{{Data Visualization}}

In [[statistics]], '''exploratory data analysis''' (EDA) is an approach of [[data analysis|analyzing]] [[data set]]s to summarize their main characteristics, often using [[statistical graphics]] and other [[data visualization]] methods. A [[statistical model]] can be used or not, but primarily EDA is for seeing what the data can tell us beyond the formal modeling and thereby contrasts with traditional hypothesis testing, in which a model is supposed to be selected before the data is seen. Exploratory data analysis has been promoted by [[John Tukey]] since 1970 to encourage statisticians to explore the data, and possibly formulate hypotheses that could lead to new data collection and experiments. EDA is different from [[Data analysis#Initial data analysis|initial data analysis (IDA)]],<ref>{{cite book |last=Chatfield |first=C. |year=1995 |title=Problem Solving: A Statistician's Guide |publisher=Chapman and Hall |isbn=978-0412606304 |edition=2nd }}</ref><ref>{{cite journal |doi=10.1371/journal.pcbi.1009819|title=Ten simple rules for initial data analysis|year=2022|last1=Baillie|first1=Mark|last2=Le Cessie|first2=Saskia|last3=Schmidt|first3=Carsten Oliver|last4=Lusa|first4=Lara|last5=Huebner|first5=Marianne|author6=Topic Group "Initial Data Analysis" of the STRATOS Initiative|journal=PLOS Computational Biology|volume=18|issue=2|pages=e1009819|pmid=35202399|pmc=8870512|bibcode=2022PLSCB..18E9819B |doi-access=free }}</ref> which focuses more narrowly on checking assumptions required for model fitting and hypothesis testing, and handling missing values and making transformations of variables as needed. EDA encompasses IDA.

==Overview==

Line 8 ⟶ 7:

Tukey defined data analysis in 1961 as: "Procedures for analyzing data, techniques for interpreting the results of such procedures, ways of planning the gathering of data to make its analysis easier, more precise or more accurate, and all the machinery and results of (mathematical) statistics which apply to analyzing data."<ref>[http://projecteuclid.org/download/pdf_1/euclid.aoms/1177704711 John Tukey-The Future of Data Analysis-July 1961]</ref>

Exploratory data analysis is an analysisa technique to analyze and investigate thea data setdataset and summarize theits main characteristics. ofA the dataset. Mainmain advantage of EDA is providing the data visualization of data after conducting the analysis.

Tukey's championing of EDA encouraged the development of [[Computational statistics|statistical computing]] packages, especially [[S (programming language)|S]] at [[Bell Labs]].<ref>{{Citation |last=Becker |first=Richard A. |title=A Brief History of S |publisher=AT&T Bell Laboratories |place=Murray Hill, New Jersey |access-date=2015-07-23 |url=http://www2.research.att.com/areas/stat/doc/94.11.ps |format=PS |archive-url=https://web.archive.org/web/20150723044213/http://www2.research.att.com/areas/stat/doc/94.11.ps |archive-date=2015-07-23 |quotation="... we wanted to be able to interact with our data, using Exploratory Data Analysis (Tukey, 1971) techniques."}}</ref> The S programming language inspired the systems [[S-PLUS]] and [[R (programming language)|R]]. This family of statistical-computing environments featured vastly improved dynamic visualization capabilities, which allowed statisticians to identify [[outlier]]s, [[trend estimation|trends]] and [[pattern recognition|patterns]] in data that merited further study.

Tukey's EDA was related to two other developments in [[statistical theory]]: [[robust statistics]] and [[nonparametric statistics]], both of which tried to reduce the sensitivity of statistical inferences to errors in formulating [[statistical model]]s. Tukey promoted the use of [[five number summary]] of numerical data—the two [[extreme value|extreme]]s ([[maximum]] and [[minimum]]), the [[median]], and the [[quartile]]s—because these median and quartiles, being functions of the [[empirical distribution function|empirical distribution]] <!-- [[statistical functional]]s (and the related [[interquartile range]] and [[range]]) -->are defined for all distributions, unlike the [[mean value|mean]] and [[standard deviation]];. moreoverMoreover, the quartiles and median are more robust to [[skewness|skewed]] or [[heavy-tailed distribution]]s than traditional summaries (the mean and standard deviation). The packages [[S (programming language)|S]], [[S-PLUS]], and [[R (programming language)|R]] included routines using [[resampling (statistics)|resampling statistics]], such as Quenouille and Tukey's [[resampling (statistics)#Jackknife|jackknife]] and [[Bradley Efron|Efron]]{{'s}} [[bootstrapping (statistics)|bootstrap]], which are nonparametric and robust (for many problems).

Exploratory data analysis, robust statistics, nonparametric statistics, and the development of statistical programming languages facilitated statisticians' work on scientific and engineering problems. Such problems included the fabrication of semiconductors and the understanding of communications networks, both of which concernedwere of interest to Bell Labs. These statistical developments, all championed by Tukey, were designed to complement the [[analytic function|analytic]] theory of [[statistical hypothesis testing|testing statistical hypotheses]], particularly the [[Pierre-Simon Laplace|Laplacian]] tradition's emphasis on [[exponential family|exponential families]].<ref>{{cite journal |title=Conversation with John W. Tukey and Elizabeth Tukey, Luisa T. Fernholz and Stephan Morgenthaler |journal=Statistical Science |volume=15 |issue=1 |year=2000 |pages=79–94 |doi=10.1214/ss/1009212675|last1=Morgenthaler |first1=Stephan |last2=Fernholz |first2=Luisa T. |doi-access=free }}</ref>

== Development ==

Line 102 ⟶ 101:

* [[Orange (software)|Orange]], an [[open-source software|open-source]] [[data mining]] and [[machine learning]] software suite.

* [[Python (programming language)|Python]], an open-source programming language widely used in data mining and machine learning.

* Matplotlib & Seaborn are the Python libraries used in todays world for EDA and Plotting/Data Visualization.(point updated: 2025)

* [[R (programming language)|R]], an open-source programming language for statistical computing and graphics. Together with Python one of the most popular languages for data science.

* [[TinkerPlots]] an EDA software for upper elementary and middle school students.