© Andrew Coles       Terms & Conditions | Privacy

Modified VWAP Methodologies
Share on Facebook
Share on Twitter
Share via e-mail
Share on Delicious
Share on Digg
Share on Google Bookmarks
Share on LiveJournal
Share on Newsvine
Share on Reddit
Share on Stumble Upon

Chart differences

Preliminary note on charting MIDAS curves

Throughout the book and his blog posts, David Hawkins has a preference for Equivolume charts. Equivolume is a method of charting developed by Richard Arms Junior. Arms bases Equivolume charting on the principle that the market is a function of volume and not time.

To accommodate this idea, price and volume in Equivolume charts are combined in one chart pane so that times of heavy trading volume are emphasized while light volume periods are deemphasized. David Hawkins prefers this method because the horizontal x-axis is proportional to cumulative volume and not time. One consequence of this is that MIDAS Support/Resistance curves and TopFinder/ BottomFinder (TB-F) curves will often plot more smoothly. An additional advantage in the case of the TB-F curves is that a linear regression extension line can be applied to price to intersect at the termination of the duration (cumulative volume) of the move required to create a TB-F curve. For the moment this will sound technical but it will be explained later. Alternatively readers can consult David Hawkins’ Chapter 7 of the book.

By contrast, Andrew Coles prefers candlestick charting which has time along the x-axis and not cumulative volume. Two advantages of using time-based candlesticks are first that their familiar reversal patterns can act as a filter for signals provided by the MIDAS system; and second, that they allow robust trade-management approaches in the setting of market stops. See for example Andrew Coles’ Chapter 1 of the book and in particular the section The MIDAS Approach as a Genuine Standalone Trading System.

It is possible to use MIDAS in chart formats beside Equivolume and conventional candlesticks, including Point-and-Figure, Three Point Break, and others.

This essay was written by Andrew Coles - no unauthorised use of this material is permitted. © Andrew Coles

Essay Two - New MIDAS Curves and New Indicators Up to the Book’s Publication: An Overview

The need for this essay

As outlined in the MIDAS orientation page here, the need for this essay grew out of a reflection after the book was published that its introduction didn’t provide sufficient clarification of how the MIDAS system had evolved during the book’s preparation and the publication of related articles. As noted on the orientation page, I (Andrew Coles) had merely referenced this evolution in the following remarks:

However, in many respects the technological changes that have affected the markets since [Levine’s] time on the hardware and software fronts mean that approaches to using the MIDAS method have inevitably evolved too, especially for contexts such as day trading and new makets. It has therefore been important to retain the basic authenticity of Levine’s teachings while allowing the approach sufficient flexibility to apply to these new areas, including the development of new MIDAS-based indicators.

Ideally this evolution would have been placed in a short conclusion to the book but the project had already reached the maximum page length. This, plus concerns over the viability of a longer introduction, prevented too much additional discussion.

The upshot was that it was sometimes asking a little much for those with less knowledge of MIDAS to appreciate to the same extent how certain ideas and techniques were changing, and with them the opportunity for new indicators. A long blog post on the previous site addressing this issue has now been destroyed along with other significant content. The aim of this second essay is to address the loss of this blog post in a more permanent format here.

Summary of blog post - types of curve

First generation curves

In the blog post, I (Andrew Coles) referred for the first time to Paul Levine’s curves (ie, the M-S/R and the TB-F) as “first generation” curves. Nothing disparaging was meant by this term. It merely reflected the fact that Levine’s curves were the first curves to be developed in the MIDAS project.

Second generation curves

In the same post, I (Andrew Coles) used the term “second generation” curves to refer to curves which in the book we called “nominal” curves. We have also called these curves “constant volume” curves. Second generation curves have a markedly different volume input to first generation curves.

As we showed in the book, second generation curves are a vital development in the MIDAS approach for two reasons. First, the use of these curves is critical in market contexts where volume is either fluctuating excessively or where there is a very large and pronounced volume trend. In such cases, first generation curves (utilising market volume) are much less dependable and often produce inaccurate results. Paul Levine was unaware of this volume impact and so never considered the possibility of new curves to tackle it (indeed, his term “porosity” was introduced to cover unexpected price behaviour around curves which we can now explain by certain volume characteristics). The second use of these curves is in markets where there is no volume - in particular, the cash foreign exchange markets. I’ll expand on both of these uses below.


back to orientation next page