Studies: Bradley Stock Model
|
The Bradley Model is a data file that is overlaid on the host chart. Click menu File | Open | Internet Services. Select the Download tab and wait for the form to fill in its dates, and then use the History drop down box to select the Bradley.d entry.

Then click the Download button.... this will put the Bradley.d file in the C:\Ensign\Day folder. Next step would be to open a chart for the $INDU.D daily file. Refresh the chart if it is not current.... those on other data feeds might have a slightly different symbol. With the INDU chart showing, click on the Study button for the study drop down list and select Overlay Chart.

Type Bradley as the Symbol, and select Line Chart as the Chart Type and Hidden as the Chart Scale. The particular example shown also has the Invert Pattern checked. Then close the study property form. When the overlay is on the chart, you can press the letter 'I' key to invert the pattern. Press 'I' key again to uninvert it.

Double clicking on the price scale will change the selection from the INDU scale to the Bradley scale. Then you can drag vertically on the Bradley scale to resize the overlay, and you can drag vertically on the chart to shift the overlay up or down relative to the INDU host.
You can adjust the overlay size and position to obtain a pleasing relationship with the host chart. Any questions? Double click the price scale again to return to the INDU scale and then you can drag the host chart leftward to you have more right side margin showing and thus be able to see the Bradley pattern in the 2007 future.
The y-axis is not fixed. The two items plot on totally different scale ranges. So I have shown you how to resize the overlay and reposition the overlay so you have a pleasing association. Ok, enjoy having the Bradley overlay as an enhancement to the Ensign Windows program.
The pattern is date specific, and you should have the overlay property of Lock Shift L/R checked so the two cannot be moved separately. You can shift the x-axis when they are locked and they both move together to stay date aligned. Dragging horizontally on the time stamps and this will tighten or widen the bar spacing so you see more or fewer bars in the view.
In 1946, Donald Bradley wrote a small paperback book entitled Stock Market Predictions. It cost a whopping $4.00 (which in 1947 was a lot of money). Llewellyn Publications from Minnesota sold very few copies. Bradley only included 2 years of data for the book from 1947-1948. The book was not mass marketed and laid in obscurity for many years.
During the transition years from pattern recognition into astrology (I have returned in earnest), the book was given to me by one of my mentors Dr. Ruth Miller of Indiana. The importance of the little pamphlet would not be revealed to me until 1987. Astrology had pulled me into its intricate web and I was writing Astro Cycles a Bi-Weekly publication featuring stock indices and commodities. The astrological data I was using came from the Astro Computing Company of San Diego, California. Neil Michelson, the owner, had become a good friend and introduced me to many aspects of astrology that I wasn't even aware they existed. When I mentioned the Bradley book and the sidereal graph, Neil volunteered to run the tests to see if it could pass rigid statistical study. Good financial data was not readily available until 1876, ten years after the Civil War. I remember this well because I was only 10 at the time.
Neil proceeded to run all years from 1876 to 1987. The results were amazing! Although it didn't work every year, statistically it was better than 65% predictable at forecasting the trend of the stock market for the year. What is truly incredible is the fact that you can do these trends decades in advance.
Why it works is an elusive answer. It is the reason I came back to pattern recognition. The chart patterns repeat and are predictable within a chaotic market environment. In order to keep it simple I found that the ratios and patterns are what makes it tradable and profitable.
The Bradley model has been programmed by our friends at Ensign Software and will now be featured as a weekly segment of my Trading Tutor newsletter.
Those of you in search of answers about what may cause this should read Dr. Al Larsen's classic book, My Electric Life. The book describes a theory of electron magnetic energy that sure makes sense. It should be in everyone's library (only $25.00).
The chart below demonstrates the Bradley Stock Market Model laid over the Dow Jones so that you can compare the astrological forecast with the markets performance in recent months.
The Bradley model is a forecast of the market based on astrological relationships. Because astrological relationships can be defined with mathematics, the Bradley forecasts can be made decades in advance.




The correlation has been excellent for the past several months. The timing of the bottom turn in November 2003 was great, as was the timing for tops in February and April 2004. The following chart shows the Bradley model for the balance of 2004, which is basically a downtrend to a turn in January 2005. The model suggests a rally from Labor Day through Election day.

It needs to be emphasized that the Bradley model does not give price levels. It denotes patterns, trends and the timing of swing tops and bottoms. The Bradley data are numbers ranging from -200 to 200. This data is rescaled and repositioned vertically on the Dow Jones Industrial chart to create the examples.

Now that 18 months have gone by, it is time for a follow up article to document the correlation of the Bradley model for 2005 with the stock market.
The Bradley model is a forecast of the market based on astrological relationships. Because astrological relationships can be defined with mathematics, the Bradley forecasts can be made decades in advance. The following chart shows how the Bradley model correlated with the stock market in 2005.

The end of 2004 and the first two months of 2005 had excellent correlation with the timing and direction of the turns in the market. However, in my opinion, it would have been difficult to trade the stock market using the Bradley model for the balance of 2005. For a portion of the summer and fall, there is better correlation with the market if the Bradley model is plotted inverted.
There are three characteristics of the Bradley forecast: time, direction, and price. Each of these will be discussed.
The primary characteristic to be extracted from the Bradley model is time. The model is basically a clock based on the motions of objects in our solar system. We readily acknowledge the influence in our lives of the daily rotation of the earth, and the monthly orbit of the moon, and the cycle of seasons from the annual orbit of the earth around the sun. But beyond those three accepted astronomical clocks, skepticism increases that life on earth is also influenced by other astronomical bodies. It is hard to accept that other astronomical bodies have any influence on the business cycles of the economy of the United States or the world. But that is what the Bradley model is attempting to show. When the timing of market turns is in synch with the Bradley forecast as was the case for the turns in December 2004 (top), January 2005 (bottom) and February 2005 (top), the forecast can be very impressive.
As is shown in the 2005 chart, the market and the Bradley forecast can be out of sync. For example, the two turns in August 2005 were well aligned for time, but out of sync for direction. The market put in a top in mid August when the forecast was for a bottom turn. And the market made a bottom swing at the end of August when the forecast was for a swing top. When this happens, the forecast is described as being 'inverted'. The timing of the turns is still well aligned or correlated. It is the direction into these turns that is inverted. Why does inversion happen? Answer: I do not know and I have not found a good answer from those who regularly work with the Bradley model. Is there a way to know in advance when inversion will happen? Answer: No. You can suspect the model is inverted when it is happening, and still take advantage of the time forecast.
The third characteristic is price, and this should have a low priority in comparison to time and direction. The Bradley data values are in the range of -200 to 200 and thus to plot the Bradley forecast, the data set has been resized and repositioned so that it shows on the INDU chart as an overlay. The Bradley curve has been stretched and shifted vertically until the curve fit nicely on the INDU chart. How the curve is shown on the chart is completely arbitrary for the convenience of seeing the Bradley curve near the market data. At times there is good correlation in comparing the size of one Bradley swing with another and seeing a similar ratio in swing amplitudes in the market. At other times, there is no price correlation. The 2005 Bradley model suggested the annual top of the market would be in July 2005. There was a swing high in the summer, but it definitely was not the top of the market for 2005. The stock market's 2005 top occurred in March with a retest of that top in November 2005.
Be sure and read the other two articles about the Bradley model to see other examples where there was better correlation between the forecast and the market than was experienced in 2005. Neither article shows the full correlation for 2004 so the 2004 chart is shown here.

This update on the Bradley model concludes by showing the forecast for 2006. Only time will tell whether the market and the forecast will be well correlated, or inverted, or down right useless as a tool for trading the stock market. The forecast does not show a price scale on purpose. Use the curve primarily for timing the turns, and then secondly for the direction into those turn dates. In general the model shows an up trend for the first half of 2006 and then a down trend into Thanksgiving. However, the forecast into the end of 2005 appears to be inverted, so the 2006 forecast may also be inverted in part or in full.

Last modified 10/23/08 2:29 PM
|