Calendar Heatmaps are an interesting alternative view of time-series data. The measured value is displayed as color mapped cells in a calendar. Calendar Heatmaps can be easily created with SAS 9.3 using just the HEATMAPPARM, SERIESPLOT and BLOCKPLOT statements in GTL and some simple data manipulation.
The example below shows the number of traffic fatalities in the US[1] during 2008. The effects of weekends and holidays can be clearly seen in this graph.
proc template; define statgraph cal_heatmap; begingraph / designheight=210 designwidth=1200; entrytitle 'Traffic Fatalities in the US during 2008'; rangeattrmap name="rmap"; range 0 - max / rangecolormodel=(white cxeeeeaa red); endrangeattrmap; rangeattrvar attrmap="rmap" attrvar=cvar var=fatalities; layout overlay / walldisplay=none xaxisopts=(display=none) yaxisopts=(display=(tickvalues) reverse=true linearopts=(tickvaluelist=(1 2 3 4 5 6 7)) ); heatmapparm x=week y=weekday colorresponse=cvar / name="heatmap" xgap=1 ygap=1; seriesplot x=wk y=wd / connectorder=xvalues break=true lineattrs=(color=black) primary=true; innermargin / align=bottom; blockplot x=prevwk block=month / valuehalign=center display=(values); endinnermargin; continuouslegend "heatmap" ; endlayout; endgraph; end; run; proc sgrender data=combined template=cal_heatmap; run;
The example above was created by generating a day of the week and week of the year value from the date value and using those columns as the X and Y values on the HEATMAPPARM. The month boundaries are computed from the input date variable and rendered using the SERIESPLOT statement and the month labels are drawn using a BLOCKPLOT statement.
When working with financial data, weekly trends as well as the values at the end of month, quarter or year can also be more easily observed on a Calendar Heatmap. The example below shows the % daily change in the Dow Jones composite index during 1988.
proc template; define statgraph cal_heatmap; begingraph / designheight=210 designwidth=1200; entrytitle 'Calendar Heatmap of % daily change of the DOW Jones composite index in 1988'; rangeattrmap name="rmap"; range min - 0 / rangecolormodel=(darkred red yellow); range 0 - max / rangecolormodel=(yellow green darkgreen); endrangeattrmap; rangeattrvar attrmap="rmap" attrvar=cvar var=percent; layout overlay / walldisplay=none xaxisopts=(display=none) yaxisopts=(display=(tickvalues) reverse=true linearopts=(tickvaluelist=(1 2 3 4 5 6 7)) ); heatmapparm x=week y=weekday colorresponse=cvar / name="heatmap" xgap=1 ygap=1; seriesplot x=wk y=wd / connectorder=xvalues break=true lineattrs=(color=black) primary=true; innermargin / align=bottom; blockplot x=prevwk block=month / valuehalign=center display=(values); endinnermargin; continuouslegend "heatmap" / valuecounthint=3; endlayout; endgraph; end; run; proc sgrender data=combined template=cal_heatmap; run;
As in the first example, the day of the week and the week of the year were computed and used for the X & Y values on the HEATMAPPARM. A RANGEATTRMAP is used with the COLORRESPONSE option to clearly represent the positive and negative changes in the Dow Jones composite index. The empty cells correspond to days where the Stock Exchange was closed.
[1] Data Source: http://www-fars.nhtsa.dot.gov
4 Comments
Will these heatmaps run in 9.2?
The HEATMAPPARM statement was added to GTL at SAS 9.3 so this code would not run at 9.2. Prior to 9.3, users have created Heatmaps using the SCATTERPLOT statement. This example illustrates how this could be done: http://support.sas.com/kb/35/156.html
For another 9.2 example, complete with a SAS program, see
http://blogs.sas.com/content/iml/2011/01/14/two-letter-initials-which-are-the-most-common/
For more on using calendar charts to visualize time series, see
http://support.sas.com/rnd/app/papers/airlinedelays.pdf
and
http://pubs.amstat.org/doi/abs/10.1198/jcgs.2011.2de (Click "Hi-res for printing")
You can also look at "Stretching the Bounds of SAS/GRAPH® Software" ... http://www2.sas.com/proceedings/sugi30/137-30.pdf.
There are also examples at http://www.albany.edu/~msz03/temps.html together with a link to daily temperature data for US and world cities (nice for practice with heat maps)
One Trackback
[...] can display the heat map with a ColorResponse rather than ColorGroup. An example of this is the Calender Heatmap posted earlier in this blog by Pratik [...]