Thursday, April 17, 2014

Star Plots



Star plots are a visual technique of portraying multivariate data. Each star equates to one observation. The star plot shown above contains plots of 16 cars. There are 9 variables to which these star plots are made from: price, mpg, 1978 repair record, 1977 repair record, headroom, rear seat room, trunk space, weight, and length.



Source URL: http://www.itl.nist.gov/div898/handbook/eda/section3/starplot.htm

Correlation Matrix


A correlation matrix is a matrix giving the correlations between all pairs of data sets.
The correlation matrix shown above is of the advice-only securities firm, Investors RouteMap and it shows the relationship between the principal bond and forex markets in addition to equity regions, sectors and select investment styles from Investors RouteMap generates recommendations.


Source URL: http://www.investors-routemap.co.uk/Guidebook_price_correlation.html

Similarity Matrix


A similarity matrix is a matrix of scores that represent the similarity between a certain amount of data points. The similarity matrix shown above is of behavioral clustering. 



Stem and Leaf Plot




Stem and leaf plots are plots that organize numerical values in terms of place value.
This stem and leaf plot shown above depicts the ages of people at a family reunion. The ages of people at the reunion are 1, 8, 9, 32, 34, 37, 45, 51, 55, and 81.



Source URL: http://www.eduplace.com/math/mhm/5/06a/

Box Plot




Box plots are diagrams that portray a set of data that is measured in maximum and minimum values, upper and lower quartiles, and a median. In the box plot shown above, the median is the horizontal line in the red box. The three intervals depict the lottery numbers that won for three different times.




Source URL: http://www.stat.yale.edu/Courses/1997-98/101/

Histogram



Histograms are bar graphs where the width of the bars are equal to the class interval and the height of the bars are proportional to the class frequencies. The histogram shown above is measuring the frequency distribution of height of 26 students.




Source URL: http://www.math.kent.edu/~honli/teaching/statistics/Chapter2/Excell_Histogram.html

Parallel Coordinate Graph


Parallel coordinate graphs are graphs that portray high-dimensional geometry with multivariate data. The parallel coordinate graph shown above is of rapper Jay-Z's career and is between 7 variables: year, word count, syllable count, polysyllabic word count, sentence count, flesch, and SMOG.


Source URL: http://staplecrops.com/data-visualization-of-jay-zs-career/