Categories

Pragmatic Capitalism

Capital for Living a More Practical Life

How to Interpret These Two Charts

« Back to Previous Page
0

Here are two charts comparing the performance of XLE and VTE over two time frames” 20 years and 2 years. Why is the last two years on the 20-year chart so different form the 2-year chart. I could understand if the shorter time scale merely opened the gap between the chart lines, but to switch the relationship? What is going on here?

Attachments:
    Marked as spam
    Posted by Lucas
    Posted on 07/31/2017 2:51 PM
    109 views
    Private answer

    The relative movements are same in both charts. The Y axis median position of the lines on a percentage scale swaps around because the denominator of the percentage is endogenous to each metric, and a different start point for the denominator is chosen in the second chart.

    Marked as spam
    Posted by Dinero
    Answered on 07/31/2017 4:53 PM
      Private answer

      Energy just beat the pants off the broad market during the early 2000s when oil prices boomed. So you’re kind of cherry picking a generous starting point by starting the energy chart where it starts….The last 2 years have been much less friendly to oil prices….

      Marked as spam
      Cullen Roche Posted by Cullen Roche
      Answered on 07/31/2017 4:57 PM
        Private answer

        Actually an addition to my first comment. They have done something much more peculiar. They have restarted the x axis at a later date but they have used the same start date for the price for both charts to calculate the percentage change.

        This illustrates Something very specific. The recent fluctuation compared to the 1999 price. Not sure how relevant that is.

        Marked as spam
        Posted by Dinero
        Answered on 07/31/2017 6:02 PM
          Private answer

          Last observation – There is not enough information to state quite what they have done there. The Short answer is the figures are not absolute values and so the position on the Y axis for each metric does not have to be consistent between the two charts even if the trends are correct.

          Marked as spam
          Posted by Dinero
          Answered on 07/31/2017 6:12 PM
            Private answer

            I don’t understand any of the responses. It is the presumably the same data. One chart is simply on a longer time scale. I could see if the shorter time scale amplified the gap between the two chart lines because the data is not so squished together. I do not understand why the two charts suggest radically different conclusions.

            Marked as spam
            Posted by Lucas
            Answered on 07/31/2017 7:57 PM
              Private answer

              You’re starting them from two different points in time. Think of it like this. Your first chart shows XLE growing from $1 to 3$ and VTSMX growing from $1 to 2$. But your second chart resets the starting point to 2015 where they’re both worth $1 to begin. And since then $1 invested in XLE has turned into $0.90 and $1 invested in VTSMX has turned into $1.10.

              Make sense?

              Marked as spam
              Cullen Roche Posted by Cullen Roche
              Answered on 07/31/2017 8:07 PM
                Private answer

                Yes on one level. But then I do not see how to use these charts in we cannot consider the 2-year chart a subset of the 20-year chart.

                Marked as spam
                Posted by Lucas
                Answered on 08/02/2017 1:28 AM
                  Private answer

                  If you look closely at chart no.1 you see that the larger growth for VTSMX staring at half way through 2015 is also there in chart no.1.

                  You ask how they could be used.
                  They could used to show that XLE has grown more in the last 17 years but VTSMX has grown more in the last 2 years. But I would not recommend it. It seems a clumsy way of doing it. Superimposing two lines on the same chart using percentages with different numerators and then compounding the confusion by resetting the datum point entirely is more confusing than enlightening, as your original post shows us.
                  That is just my opinion.

                  In summary Cullen’s repply is correct.

                  Marked as spam
                  Posted by Dinero
                  Answered on 08/02/2017 9:14 AM
                    Private answer

                    I’ve seen this kind of chart before at Yahoo. 2yr is not a fractal of the 20yr because the rate-of-change return is only calculated from the leftmost date showing on any particular zoom in or zoom out. If it was calculated from the beginning of the data regardless of the zoom setting, then it would be fractal and match.

                    Marked as spam
                    Posted by MachineGhost
                    Answered on 08/02/2017 8:08 PM
                      « Back to Previous Page