Website Update

In the next couple of days I will have the code/scripts ready to batch create 3D contour maps in Three.js for all the thousands of currently monitored markets and stocks. They will then be updated on a weekly basis.

Also, with the help of D3.js in Action by Elijah Meeks I will change and improve the current D3 data graph. If there are any suggestions please leave a comment. My plan is to have the data available for 15 or more windows via a legend/check box/drop-down menu; the user will be able to select/choose which windows he/she wants plotted on the D3 graph. For examples and potential ideas: D3 Examples. I think this one would be good for the site: UK Temperature History.

LPPL Oscillations as Vernadsky’s “Pressure of Life”

The H,Q derivative approach taken by The Bubble Index was developed from the paper, Generalized q-Analysis of Log-Periodicity: Applications to Critical Ruptures by Wei-Xing Zhou and Didier Sornette. This approach has its origin in studying the critical ruptures arising in pressure storage tanks. Pressure and LPPL Oscillations are connected. This implies the existence of some sort of pressure build-up which occurs in financial markets which create LPPL Oscillations. What is this pressure mechanism?

I wonder if The Bubble Index (and LPPL Oscillations in general), might be measuring directly/indirectly the Vernadskian “pressure of life” exerted by individual social groups of H. sapiens on the biosphere. This scale-free “pressure of life” exerted by these groups of H. sapiens on the environment builds over time, until it reaches a critical point when the system ruptures. A crash in wealth and a shock to economic growth occurs to relieve the pressure which inevitably results in a depression/recession as the “pressure of life” exerted by H. sapiens begins to find a new state.

Growth is a double edged sword – creating pressure on the biosphere while increasing the prosperity of H. sapiens in aggregate. However, if this pressure can now be directly/indirectly measured, the interaction between H. sapiens and the biosphere can be engineered. It is possible that the prosperity of H. sapiens can be designed to increase with a simultaneous reduction in pressure exerted on the biosphere.

The Bubble Index Contour: DJIA (1871 – 2015)

The results for The Bubble Index Contour: DJIA from 1871 to 2015 are presented below. There are several features worth mentioning.

Overall Trend

The first is the overall trend toward larger absolute values as time progresses towards the present.

23,200 Day Window Growing Pressure

The second most striking feature, which is most relevant to current investment strategies, is the large pressure formation centred on the 23,200 day window. The values in the graph were interpolated with a spline, so there are some artefacts produced. They falsely seem to indicate that the pressure has peaked in the 23,200 window, however, upon inspecting the raw data, this time window has not yet peaked. However, the derivative is close to and fast approaching zero. This would seem to indicate a phase transition of the largest magnitude yet observed will occur in the near future.

2008-9 Pressure Peaks

The pressure peak below and slightly to the left of the current 23,200 pressure growth corresponds to the bottom of the 2008-9 crash, where the DJIA reach a low of 6,443. This pressure point has a peak which occurred in the 15,000 day window.

The pressure peak below and slightly to the left of the aforementioned 15,000 day window peak was the peak in the 9,500 day window which corresponded to the peak value of 14,200 in the DJIA in 2007.

Other images:

djia-17  djia-2 djia-3 djia-4 djia-7   djia-8 djia-9 djia-10 djia-11 djia-12 djia-13 djia-14 djia-15 djia-16

TNX Update

With the upcoming FOMC meeting, it is worth revisiting the CBOE 10-Year Gov’t Rate (TNX). It appears that the 8,100 day window, which recently peaked is trying to find a direction. Will the bubble continue? In the actual data files, the 8,200 day window peaked on February 13; the 8,100 day window also peaked on February 13 — only to peak again on February 27. This indicates that we could be near a phase transition in the 10-Year rate. I am not yet sure what this graph implies. This is a very long term signal, since 8,200 days is about 33 business years. Remember that not all bubbles are followed by a crash.

tnx_10 tnx_11 TNX_12 TNX_13 TNX_14

WebGL and 3D Contours

After a long night of learning how to code Three.js, I have had some success in rendering the contours created by The Bubble Index into nice WebGL displays. You can find them in the 3D Contours Page on this site.  The process that created these consisted of several steps. First, I exported a contour 3D grid mesh as a DXF file from 3DField. Then I used JNetCAD to convert this into a WebGL json file.  JNetCAD describes itself as: JNetCAD is a simple Java GUI for importing and exporting several CAD formats. I agree, it was simple and effective for my purposes. From this point, it was now about coding in Three.js. Three.js has some really nice examples to learn from. It gives users the potential to create some amazing visual displays. And after some debugging and tweaks, I ended up with a nice display.

3D Contour Map: TNX (1962 – 2015)

The 3D Contour of The Bubble Index Contour: TNX (1962 – 2015) displays an important prediction. As you can see from the graphs and contours, it appears to have peaked in the window range 7,200 – 8,200. This would indicate that the bubble in the US Gov’t 10 – Year Bond recently ended. That is cautioned by an unknown factor — that there is not a bubble forming on a much larger time window than I was able to calculate (13,200 days +).


3D Contour Videos

In order for viewers to have a better visual perspective of the 3D Contours, I have uploaded the videos below to YouTube. These are a test of potential routes for how the website could develop and serve to help me decided how to proceed in the future. Eventually, I think I will upload versions which contain narration and investing ideas based on the graphs. Once I figure out how to map the date axis from integers to Dates that should improve the visual experience. In addition, I hope to figure out and be able to overlay the underlying time series for the 3D Contour in the foreground and or background panel.