Cash in, cash out graph

David Afshartous writes:

I thought this graph [from Ed Easterling] might be good for your blog.

The 71 outlined squares show the main story, and the regions of the graph present the information nicely.

Looks like the bins for the color coding are not of equal size and of course the end bins are unbounded. Might be interesting to graph the distribution of the actual data for the 71 outlined squares. In addition, I assume that each period begins on Jan 1 so data size could be naturally increased by looking at intervals that start on June 1 as well (where the limit of this process would be to have it at the granularity of one day; while it most likely wouldn’t make much difference, I’ve seen some graphs before where 1 year returns can be quite sensitive to starting date, etc).

I agree that (a) the graph could be improved in small ways–in particular, adding half-year data seems like a great idea–and (b) it’s a wonderful, wonderful graph as is. And the NYT graphics people added some nice touches such as the gray (rather than white) background and the thin white lines to separate the decades.

On a (slightly) more substantive note, I don’t think growth-adjusted-for-inflation is the best benchmark. Instead of growth minus inflation, I’d like to see growth minus the default interest rate you could get from a savings account or T-bill or something like that. Lots of possibilities here.

7 thoughts on “Cash in, cash out graph

  1. AARRRGGH! Hasn't NYT figured out how to make colorblind friendly graphics yet? Red and green splotches of the same saturation against a neutral background are exactly what should not be used. Although this looks like it might be informative, I cannot tell the positives from the negatives – making the graphic mostly useless.

  2. yeah, it's pretty bad

    For all the trichromats out there, go to Vischeck, enter the URL of the NY Times story and click for a Deuteranope simulation (red/green color blind)

  3. I e-mailed Ed Easterling and he was kind enough to send me the data for the 20 year horizons. Here are a couple of plots: http://biostat.mc.vanderbilt.edu/wiki/pub/Main/Da

    Also, some summary stats:
    Min. 1st Qu. Median Mean 3rd Qu. Max.
    -2.000 0.800 4.100 3.404 5.450 8.400

    15.3% of the returns are below zero.

    These data might be used for justification for taking the traditional guaranteed annuity option in a 403b, e.g., TIAA-CREF currently offers 3.35% in employer plans which isn't too far off the median above. However, perhaps not a good comparison since these are real returns which would have been greater if held in a 403b since there wouldn't have been taxes on dividends, etc., making the guaranteed option look worse.

    I checked out Ed's web site and there lots of other interesting graphs:
    http://www.crestmontresearch.com/content/market.h

    david

  4. I like the graph (red/green snafu aside). Anything that can help people understand variance.

    The main problem I see is that this overstates the variance that most investors will see. On the other hand, it's very easy to interpret.

    A more realistic estimate would assume someone invests some fraction of their income every year for a period of years and that income grows over time. All of this reduces the final variance, though of course, the initial investments get more compounding.

  5. Bob, perhaps the right mix of colours?

    Although it would exclude some viewers, using a weighted mix of colours (weighted by fraction of incomes) might be an interesting exercise.

    And this is a selfish comment as I am trying to do the same with priors and likleihoods and have just started looking at colorspace and RColorBrewer in R but would rather find out someone has already figured out such schemes and clearly written about them in a free accessible format – yes selfish.

    K?

  6. "I'd like to see growth minus the default interest rate you could get from a savings account or T-bill or something like that. Lots of possibilities here."

    Could've made for a good interactive graphic where you can adjust the comparison.

Comments are closed.