Monday, December 13, 2010

The Dow Jones Industrial Average As Art

LATEST UPDATE OF CHERNOFF FACES: 31 JAN 2011.
According to the Chernoff faces, the DJIA (and by implication the general market, with a bias towards big cap bluechips) remains still much the same. Size of face is proportional to valuation +/- % and if you are looking for undervaluation, look for the small faces. e.g. Alcoa and BAC. But then there are reasons why BAC is undervalued. Alcoa is not so squinty-eyed anymore, due to a combination of improvements in its Expected EPS, volatility and Beta.
Chernoff faces of the DJIA as at 31 January 2011
Chernoff faces circa 15 Dec 2010


Above is a Chernoff Face glyphplot of the DJIA stocks. For explanation of Chernoff Faces see my post at http://www.technifundamentals.com/2010/12/merry-xmas-edition-fun-with-statistics.html.
The 3D visualizations below are derived from the 30 constituent stocks of the DJIA, and their fundamentals e.g. P/E, Beta, 12-month Return, Earnings Per Share Forecast, Earnings per Share Actual etc - 15 selected variables in all. Graphs were drawn with Matlab and edited with SnagIt. Variables are first standardized into ZSCORE where the data is centered and scaled by dividing with the standard deviation.
STEM3: STEM3(Z) plots the discrete surface Z as stems from the xy-plane terminated with circles for the data value.
WATERFALL: WATERFALL(...) is the same as MESH(...) except that the column lines of the mesh are not drawn - thus producing a "waterfall" plot. AREA: AREA(X,Y) produces a stacked area plot suitable for showing the contributions of various components to a whole. BAR3: BAR3(Y,Z) draws the columns of the M-by-N matrix Z as vertical 3-D bars. The vector Y must be monotonically increasing or decreasing
CONTOURF: CONTOUR(Z) is a contour plot of matrix Z treating the values in Z as heights above a plane CONTOURF(...) is the same as CONTOUR(...) except that the areas between contours are filled with colors according to the Z-value for each level. Contour regions with data values at or above a given level are filled with the color that maps to the interval
PCOLOR: PCOLOR(C) is a pseudocolor or "checkerboard" plot of matrix C. The values of the elements of C specify the color in each cell of the plot

PLOTMATRIX: PLOTMATRIX(X,Y) scatter plots the columns of X against the columns of Y. If X is P-by-M and Y is P-by-N, PLOTMATRIX will produce a N-by-M matrix of axes. PLOTMATRIX(Y) is the same as PLOTMATRIX(Y,Y) except that the diagonal will be replaced by HIST(Y(:,i)
PRINCOMP
Above: A beautiful representation of ValuEngine's model variables applied to the thirty DJIA components. Model variables were first standardized by converting to Zscores and Principal Component Analysis was done on the selected 15 variables. What you see is a 3D surface and contour visualization of the coefficients of each Principal Component.

No comments:

Post a Comment