Visualizing the Fundamental Data on 400 Stocks over 80 Quarters

It is relatively easy to visualize the aggregated statistics over many periods, e.g. by means of the boxplot series. However, it may be challenging if you want to have a simultaneous look at every element for all time periods. We propose to do it by means of an animated 3D-scatterplot.

Previously we have shown how to visualize the market summary w.r.t. to the stock fundamental data. Such overview shall be sufficient for a macrotrader but is definitely insufficient for a stock picker, who needs to track the dynamics of single stocks, compared to each other.

Additionally, there is a problem of missing data. It is not especially acute for a macro-view on market (it is enough to have data on many but not necessarily on all stocks in order to grasp a measure of central tendency) but for a single stock it might be crucial.
But how to visualize the fundamental dynamics of several hundred stocks over 80 quarters?!

A naive way may be as follows:

ts.plot(t(dataMatrix[10:20,]), col=seq(1:10), lwd=seq(1,10),  gpars=list(xaxt="n")) 
axis(1, at=1:length(dtFrom), tick=1:length(dtFrom), labels=dtFrom, las=2) 

If the dataMatrix contains NA where the data are missing, ts.plot() will draw them correctly. Since it is very hard (at least for me) to grasp an intermittent chart, I additionally variate the line thicknesses. A more advanced approach might be to fill the gaps with thin dotted lines but in either case one can hardly put more than 10 graphs on a single chart.

Alternatively, one can make an animated chart as follows

#xxx - X-coordinates (stock order number)
#yyy - Y-coordinates (quarters)
#zzz - considered fundamental value 
s3d <- scatterplot3d(xxx,yyy,zzz)  
for(pp in 1:N_STOCKS)
    idxI = intersect(which(xxx==pp), which(!
    s3d$points(xxx[idxI],yyy[idxI],zzz[idxI], type="o", col="blue", lwd=2)
    s3d$points(xxx[idxI],yyy[idxI],zzz[idxI], type="o", col="green", lwd=2)
    s3d$points(xxx[idxI],yyy[idxI],zzz[idxI], type="o", col="black", lwd=1)

In this example we considered the equity-to-asset ratio (Eigenkapitalquote) and we see that it is pretty volatile from quarter to quarter... which means that the data from might be inconsistent, since this property is expected to be somewhat more stable.

Like this post and wanna learn more? Have a look at Knowledge rather than Hope: A Book for Retail Investors and Mathematical Finance Students

FinViz - an advanced stock screener (both for technical and fundamental traders)

Leave a Reply

Your email address will not be published. Required fields are marked *