| XLMidas gives you the tools you need to implement Dr. Paul Levine's amazing Market Interpretation/Data Analysis System. I've got say up right up front that MIDAS is one of my favorites . I wouldn't trade without MIDAS. It's that good! |
Due to changes at Yahoo! and need for updating the software XLTrader is not available at this time. Please check back later!
XLMidas gives you the tools you need to implement Dr. Paul Levine's amazing Market Interpretation/Data Analysis System. I've got say up right up front that MIDAS ranks as one of my favorite technical analysis techniques.
If you are not already familiar with the MIDAS method, then please take a moment to read my introductory article in order to familiarize yourself with the method and terminology.
At XLTrader our goal is to make technical analysis as easy as possible. In most cases all that is required of the analyst is to cut-and-paste the historical price information (O-H-L-C-V) into the data page of the HLC_Chart.xltm template then click on a graph sheet where the MIDAS toolbar is accessible.
Once the price bars are scaled the way the analyst wants them, then it is a simple matter to click the desired tool on the toolbar and follow the prompts. The macro will ask you to click on a point (H, L or Midpoint) and confirm your selection. You are then taken to the data page and asked to choose a column from a drop-box in which to put the calculations. Press OK then BOOM there's the curve to the chart. Its that easy!
But what if you are a power-user and you want more control? Perhaps you even want to write your own macros. NO PROBLEM! The MIDAS functions are all readily accessible.
MIDAS S/R Curve
The heart of the MIDAS method is the S/R curve which of course is short for Support and Resistance. This function is accessible by clicking on the crowns icon in the tool bar.
The user is prompted to choose a point at which to anchor the S/R curve. Don't worry if you don't get the right one because using the arrow icons next to the crowns you can easily shift your S/R curve left or right as necessary. By holding down the shift key while you are clicking the arrow, you can move the curve a long way fast. Its not "click-n-drag easy" but close.
The MIDAS articles talked about using the midpoint (H+L)/2 in the calculations. Our XLMidas implementation is a little more versatile. You can choose the H or L and the midpoint will be calculated for you. Per Levine's instructions, Granville's On Balance Volume (OBV), based upon the midpoint, is plotted on the secondary axis as shown on the figure above.
MIDAS Topfinder
XLMidas also comes with a topfinder/bottomfinder algorithm. It is triggered by clicking on the toolbar icon with red price bars. The arrow icons next to the topfinder icon are used to adjust the volume of the pull-back (to adjust the "C" parameter as defined in Levine's articles).
When the market takes off (plummets) it is amazing how topfinder (bottomfinder) can pick the ending price and accurately estimate the number of days away it will be.
Another way topfinder can be used is as a trailing stoploss trigger. When the market has been moving up sharply and begins to correct, if it breaks through the topfinder curve, chances are very good that it will eventually fall to S/R. Likewise if there is a sell-off underway and price puts in a bottom and in so doing breaks through the bottomfinder curve chances are very good it will reach S/R. What a great tool to having your TA bag of tricks!
Parallel Offset
You might be wondering if you overlooked something in Levine's articles. The "Parallel Offset" function was not discussed there because it is my own MIDAS extension.
After years of using MIDAS and really loving how great the S/R curves were for determining theoretical support levels I realized something was missing. I needed a way to know definitively when a pullback to support would begin. After trying many different ideas I found what I was looking for and I call it "Parallel Offset"
Midas Moving Averages
There are two moving averages in the tool set. You may be familiar with the New one which is the elastic volume weighted moving average (eVWMA) that was detailed in an article by Christian Fries in the June 2001 issue of TASC magazine. The second is our invention. It is based on the standard MIDAS SR formula but automatically resets when price meets support in an uptrend or resistance in a down trend. eVWMA is very smooth and responsive. The Midas MA has potential for use in automated quantitative systems.
Midas Price Distribution Function (PDF)
What would the the distribution look like were one to launch a Midas S/R curve at every bar? That is the question that precipitated the development of the Midas PDF function. To my knowledge Midas PDF is yet another XLTrader exclusive and we're (me, myself and I that is) still in the early stages of exploring its possibilities.
It is not practicable to launch a Midas S/R curve at every bar. The resulting cacophony of curves would likely resemble a chaotic completely indecipherable mess. But its easy for a computer to calculate the current resulting distribution of Midas SR prices and that is exactly what the Midas PDF is. While undoubtedly curves occurring at the paradigm shifts (CIT's) are the most important, computing them all may prove valuable in helping to ascertain how effective those key curves will be when tested.
Interpretation
You can calculate the Midas PDF based on any number of bins and either a specified look-back window or the entire history. Either way what typically results is a distribution shaped like a U, the low of which represents major support and the high major resistance. If the current price ((H+L)/2) is somewhere in the the trough, then the market is in congestion or cycle mode and likely to bounce between extremes. If the current price has broken through support (or resistance) then it is in trend mode. |
adjustable parameters
Once plotted, the number of bins used in the analysis is stored in row 1 on the Data sheet. You can change this at any time. You can adjust the baseline of the Midas PDF histogram using the arrow icons associated with the Midas S/R tool. This changes the number in row 2 which defaults to the chart X axis maximum. The number in row 3 represents the width of the PDF in X axis units and defaults to 1/5th of the total. You can also shift the distribution vertically using the up/down icons associated with the Midas Top/Bottom finder tool. This changes the number in row 4 on the Data sheet. You may be asking: Why would I want to shift the distribution vertically… would that not invalidate the results? Yes but…
Long time users of Midas have discovered that there is typically a consistent offset for a given ticker which defines a channel in which prices trade. If you know what this offset is for your market and the market is in trend mode, then you can shift your Midas PDF up or down by that amount and the see where to anticipate support or resistance. In effect the Midas PDF can be used in this way to indicate the channel boundaries we typically associate with a parallel offset curve.
Mobility Oscillator and Price Distribution Function (PDF)
In the February 1996 issue of TASC magazine Mel Weidner Ph.D wrote an article titled "Gauging Mobility with Price Distributions". Later Weidner penned a chapter in the book: Computerized Trading that was edited by Mark Jurik. His thesis is that the price distribution function (PDF), which analyzes the distribution of prices over a look-back period, is useful for predicting price mobility. Further to that end, he developed the Mobility Oscillator which takes into account the current price location within the PDF. At the peak of the PDF, there is low mobility (congestion… resistance to price movement) at the other extremes, there is high mobility indicating that price can more easily move higher (or lower) but ironically, is also more likely to change directions.
Using the MOB osc. indicator module you can plot the price distribution function (PDF) for the entire history or plot the price distribution for a specified look-back period. (You can elect to plot the mobility oscillator when you check the option for specifying a look-back period on the userform)
PDF Usage
You must specify: the number of bins into which the history gets partition; whether or not you want a volume weighted distribution and the output column on the Data sheet. Once the run button is pressed, you are asked to specify two colors, the first being the histogram bar color and the second, the spline curve fit of the histogram distribution.
Once the histogram has been overlaid, you can adjust the baseline using the arrow icons typically associated with the Midas S/R tool. Pressing that icon changes the number in row 2 on the Data sheet which is initally the X axis maximum. Should you decide to change the number of bins then edit the number in row 1 and if you decide to not use volume weighting, change the value in row 4 to false (blank or 0 not 1). Row 3 contains the width of the PDF histogram in X axis divisions which defaults of 1/5th of the total.
Mobility Oscillator Usage
In an attempt to boil the pdf down into a single number, Weidner developed an oscillator he calls the Mobility Oscillator (MO). MO considers the current position of price: (H+L)/2 in relation to the peak value of the histogram. Right at the peak, there is low mobility and the oscillator is zero. At the other extremes there is high mobility. If the current price is below the peak the mobility is negative, likewise if the price is above the peak then there is high positive mobility.
While divergences in the mobility oscillator may be a precursor to a change in trend generally speaking the MO is a noisy and erratic indicator and therefore the user may find it to be of limited use. However another indicator can be made by integrating the MO. By doing that one arrives at a smooth and responsive trend indicator. Both the mobility oscillator and its integration are plotted on the secondary axis.
MMulti-Timeframe Pivots
XLMidas is generally comprised of tools and techniques for finding support and resistance. One popular method for determining support and resistance are the so-called trader's pivots. Although volume is not part of their calculation, XLMidas is the logical home for this capability. Pivots are calculated based on these well known formulas:
R3 = H + 2*(PP - L)
R2 = PP + (H - L)
R1 = 2*PP - L
PP = (H + L + C)/3
S1 = 2*PP - H
S2 = PP - (H - L)
S3 = L - 2*(H - PP)
where: H is the period High
L is the period Low
C is the period Close
S stands for Support
R stands for Resistance
Pivots are calculated for these timeframes:
- day
- week
- month
- quarter
- semester
- annual
All pivots are calculated for the previous period and are not inclusive of data from the period end to the current date. For instance, if the year is 2011, then the annual pivots are calculated for the period beginning Jan 1, 2010 and ending Dec 31 2010. If the Month is May then the Monthly pivots are calculated for dates in April and Quarter pivots are calculated for dates beginning January 1 ending April 30.
All pivots are plotted on a histogram. Each bar of the histogram extends back to the start of the period for which it is calculated. If there is not enough data to calculate say… the annual pivots then those are skipped.
Interpretation
Where there is a confluence of pivots, regardless of rank (R1, R2, S3 etc.) there exists a high probability trade opportunity. When these levels coincide with support and resistance levels calculated using other methods, then they should be given extra credence.
how to use
To apply the multi-timeframe pivots simply click the icon, specify the output column on the Data sheet and choose a color. Once overlaid on the chart, the baseline of the histogram can be adjusted using the left and right arrow icons on the tool bar that are typically associated with the Midas S/R tool. When additional data is downloaded, the pivots will automatically be updated.
Midas Price vs. Cumulative Volume Template
While you can still use the midas toolset with any of the other XLTrader templates, XLMidas includes its own template which has Cumulative Volume instead of time on the X axis. It also has a second chart showing price vs. OBV so you can quickly see if price is acting correctly or is too high (which presages a selloff) for the OBV or too low (which presages a rally)