[Ed. note: we are republishing the original Frank Roellinger article on the Modified Ned Davis Method with an up-to-date performance chart]

Introduction and Disclaimer

An earlier version of this article was submitted to Seeking Alpha but was rejected because it contained no “in-depth, fundamental analysis”. A brief search of the SA website disclosed references to fundamentally based articles on the Acting Man website. So I sent Pater a note claiming knowledge of a fairly accurate, purely mechanical method to identify significant long-term turning points in the market. I offered to tell him a few days in advance the target point of the next signal. He graciously offered to publish my article here. If it can be arranged, I will post its buy and sell signals as they occur in the future. This information is for educational and entertainment purposes only, it will never be a recommendation to buy or sell anything. But I believe that it will prove interesting to consider and watch over time.

 

Method Description

My method is similar to the so-called 4 percent method on the Value Line Geometric index, as first published by Ned Davis in the early 1980s. For those unfamiliar with this index, it is described here:

http://en.wikipedia.org/wiki/Value_Line_Composite_Index

Davis’s algorithm simply bought long any 4% or greater move up in the weekly closes of this index, and sold and went short on any 4% or greater move down in the weekly closes. His algorithm captured a good portion of every major move up or down, but, as is typical of a trend following method, suffered a number of whipsaw losses, primarily due to false sell signals. I tried a number of ways to reduce these and found two that worked well.

The first added feature uses a trend line. By dynamically constructing this line and deferring action until the market penetrated it, the method reduced whipsaw losses significantly, with little effect on total returns. I thus added this feature to the method.

The second improvement concerns short sales. Davis shorted on all sell signals. Unfortunately, most of these short sales did not end profitably. Using market breadth (advancing and declining issues on the NYSE) was found to better identify conditions for a short sale. My first pass modification identified every major downturn since 1961, except the plunge on 9/11/2001 on occasion of the WTC attack (which was hardly an economically based event), and prevented shorting of many of the smaller corrections. I added this feature to the method, unchanged from my first attempt.

 

Data Series

The farther an index moves (in percentage terms), the better it is suited for trend following. Davis probably was aware of this, and chose the Value Line Geometric index because it did move farther than other indexes available at the time, such as the Dow Jones Industrial Average and the S&P 500. Many small-cap indexes exhibit this tendency to make greater moves than their larger cap cousins, and are the basis for ETFs and futures contracts. Currently there are no ETFs or futures contracts based on the Value Line index, so an alternative had to be chosen.

The data series used by the method begins with the Value Line Geometric index, because there exist no readily available small-cap indexes prior to its introduction. But conversion to another index was required before the Value Line index fell out of favor. The Russell 2000 index is currently quite popular, so a continuous index was created by using the Value Line index until the Russell index was introduced in 1979, and then splicing the Russell index onto the Value Line index.

 

Testing

The algorithm is controlled by three parameters. Two of these are buy/sell thresholds, and the third is the slope of the trend line. The algorithm could be optimized via backtesting, but a ‘forward test’ is actually far better. The typical automated trading method uses (or at least they did, for a very long time) a large number of parameters and the method is then optimized by backtesting on a great deal of historical data to determine the best set of parameter values to use. These methods work fabulously on past data when thus optimized, but often they soon begin to fail on future (out of sample) data. A forward test, where results are recorded entirely on data that the method has not used for optimization, provides a far better indication of how a method will perform in the future.

The forward test began by “training” the algorithm on the S&P 500 data from 1942 through 1960: all possible combinations of the three parameters were applied by the algorithm to the S&P 500 data, and the values that produced the best results were chosen. These parameter values then were used for the first trade on the Value Line/Russell 2000 data series, and the result of that trade was recorded as the first result of the forward test.

For the next trade, the algorithm was trained by running the Value Line/Russell 2000 data series from its beginning to the end of the first trade, with all possible combinations of the three parameters, and choosing the values that produced the best results. These parameter values then were used for the next trade on the data series, and that result was recorded as well. This process was continued for all of the data in the series, beginning each training session with the start of the data series and ending at the end of the last trade.

 

Results

The results are quite impressive: close to a 14% average gain per year since 1960 was achieved, with a maximum drawdown on closed trades of about 26%. Dividends and money market interest would boost this annual gain to over 15%. For comparison, a buy and hold strategy using the S&P 500 (excluding dividends) averaged only 6.5% per year over the same period with a maximum drawdown greater than 40%.

The accompanying chart shows the results. The upper (white) line is the method; the lower (yellow) line is the Value Line/Russell 2000 index. For comparison, the S&P 500 is shown as the green line. These are results through 11/15/2013. In this forward test the method traded about 3-4 times per year, winning on 54% of all trades. Its win/loss or payoff ratio is 3.98.

 

The return produced by the Modified Davis Method compared to the S&P 500 index and the combined Value Line/Russell 2000 Index (Value Line before 1979, Russell thereafter) – click to enlarge.

 

Risk of Ruin

The risk of ruin for this method, as defined in this article:

The Risk of Ruin Tables You Should Know

is quite low. To compute the risk of ruin value we need this information:

Risk per Trade

Payoff Ratio

Win Ratio

The method’s Risk per Trade is difficult to estimate. Its average loss is about 2.8%, but its maximum loss has been 7.25%. In 190 trades it had only 3 losses greater than 6%. We must interpolate between the 10% and 5% Risk per Trade tables.

Rounding the Payoff Ratio to 4:1 and the Win Ratio to 55%, the 10% table (10% of capital at risk per trade) gives a Risk of Ruin of .0438 and using the 5% table, the Risk of Ruin amounts to zero. So the method’s ‘actuarial’ Risk of Ruin is greater than zero, but it is obviously quite low.

 

Current Status, Future Results

Currently the MND Method is 100% long the Russell 2000 Index. If it can be arranged, I will be posting future trades here as soon as possible after the close on the day they occur, and usually hopefully before the market open of the next trading day. Stay tuned if you’re interested.

 

Positions Held

Disclosure: I and members of my family at present own shares of the IWM ETF and are long Russell 2000 mini futures contracts. We will continue to buy and sell these positions as the method dictates.

 

Addendum

 

Frank Roellinger is a retired software engineer who worked for a major computer manufacturer for nearly 34 years. He has been an avid follower of markets for more than 30 years, with a strong preference for technical over fundamental analysis. After observing the results of many different ways to approach markets, he settled upon long-term trend following. He once hoped to develop a mathematically-based model of the stock market, but now sees little point in doing that, as his Modified Davis Method works as well as any other purely mechanical method that he thought he might ever find.

 

 
 

 
 

Dear Readers!

You may have noticed that our so-called “semiannual” funding drive, which started sometime in the summer if memory serves, has seamlessly segued into the winter. In fact, the year is almost over! We assure you this is not merely evidence of our chutzpa; rather, it is indicative of the fact that ad income still needs to be supplemented in order to support upkeep of the site. Naturally, the traditional benefits that can be spontaneously triggered by donations to this site remain operative regardless of the season - ranging from a boost to general well-being/happiness (inter alia featuring improved sleep & appetite), children including you in their songs, up to the likely allotment of privileges in the afterlife, etc., etc., but the Christmas season is probably an especially propitious time to cross our palms with silver. A special thank you to all readers who have already chipped in, your generosity is greatly appreciated. Regardless of that, we are honored by everybody's readership and hope we have managed to add a little value to your life.

   

Bitcoin address: 1DRkVzUmkGaz9xAP81us86zzxh5VMEhNke

   
 

4 Responses to “The Modified Davis Method”

  • rodney:

    Continuous walk-forward optimization?

  • Frank Roellinger:

    Intriguing thought. It probably would work on commodities, but they present special problems for testing, e.g. discontinuities at contract expiration times, and locking limits. I’ve dabbled a bit in commodities, once tracking several over about 6 months using Russell Sands’ version of the Turtles method, and found it to be profitable. But it’s a whole different world in commodities and I’ve never taken the time to investigate it fully. Historically, profitable trends have come often enough in the small-cap indexes for my tastes, and the breadth phenomenon does give them an edge. There are other measures unique to stocks (not mentioned in the article) that I have found do give some useful clues about what is going on and what is likely to happen next.

  • SavvyGuy:

    It might be interesting to see how this algorithm works on different markets e.g. soybeans, bonds, metals, crude oil, etc. Even though the market breadth parameter would not be applicable to these markets, such additional testing could help improve the trend-following behavior of the algorithm.

Your comment:

You must be logged in to post a comment.

Most read in the last 20 days:

  • The Biggest Stock Market Crashes Tend to Happen in October
      October is the Most Dangerous Month The prospect of steep market declines worries investors – and the month of October has a particularly bad reputation in this respect.   Bad juju month: Statistically, October is actually not the worst month on average – but it is home to several of history's most memorable crashes, including the largest ever one-day decline on Wall Street. A few things worth noting about 1987: 1. the crash did not presage a recession. 2. its...
  • Canada: Risks of a Parliamentary Democracy
      A Vulnerable System Parliamentary democracy is vulnerable to the extremely dangerous possibility that someone with very little voter support can rise to the top layer of government. All one apparently has to do is to be enough of a populist to get elected by ghetto dwellers.   Economist and philosopher Hans-Hermann Hoppe dissects democracy in his book Democracy, the God that Failed, which shines a light on the system's grave deficiencies with respect to guarding liberty. As...
  • Federal Reserve President Kashkari’s Masterful Distractions
      The True Believer How is it that seemingly intelligent people, of apparent sound mind and rational thought, can stray so far off the beam?  How come there are certain professions that reward their practitioners for their failures? The central banking and monetary policy vocation rings the bell on both accounts.  Today we offer a brief case study in this regard.   Minneapolis Fed president Neel Kashkari attacking a block of wood with great zeal. [PT] Photo credit: Linda Davidson...
  • Thoughtful Disagreement with Ted Butler
      Too Big to Fail?   Dear Mr. Butler, in your article of 2 October, entitled Thoughtful Disagreement, you say:   “Someone will come up with the thoughtful disagreement that makes the body of my premise invalid or the price of silver will validate the premise by exploding.”   Ted Butler – we first became aware of Mr. Butler in 1998, and as far as we know, he has been making the bullish case for silver ever since. Back in the late 90s this was actually a...
  • Donald Trump: Warmonger-in-Chief
      Cryptic Pronouncements If a world conflagration, God forbid, should break out during the Trump Administration, its genesis will not be too hard to discover: the thin-skinned, immature, shallow, doofus who currently resides in the Oval Office!   The commander-in-chief - a potential source of radiation?   This past week, the Donald has continued his bellicose talk with both veiled and explicit threats against purported American adversaries throughout the world.  In...
  • Precious Metals Supply and Demand Report
      Fat-Boy Waves The prices of the metals dropped $17 and $0.35, and the gold-silver ratio rose to 77.  A look at the chart of either metal shows that a downtrend in prices (i.e. uptrend in the dollar) that began in mid-April reversed in mid-July. Then the prices began rising (i.e. dollar began falling). But that move ended September 8.   Stars of the most popular global market sitcoms, widely suspected of being “gold wave-makers”. From left to right: Auntie Janet...
  • The Donald Can’t Stop It
      Divine Powers The Dow’s march onward and upward toward 30,000 continues without a pause.  New all-time highs are notched practically every day.  Despite Thursday’s 31-point pullback, the Dow is up over 15.5 percent year-to-date.  What a remarkable time to be alive.   The DJIA keeps surging... but it is running on fumes (US money supply growth is disappearing rapidly). The president loves this and has decided to “own” the market by gushing about its record run. During...
  • 1987, 1997, 2007... Just How Crash-Prone are Years Ending in 7?
      Bad Reputation Years ending in 7, such as the current year 2017, have a bad reputation among stock market participants. Large price declines tend to occur quite frequently in these years.   Sliding down the steep slope of the cursed year. [PT]   Just think of 1987, the year in which the largest one-day decline in the US stock market in history took place:  the Dow Jones Industrial Average plunged by 22.61 percent in a single trading day. Or recall the year 2007,...
  • Stocks Up and Yields Down – Precious Metals Supply & Demand
      Where the Good Things Go Many gold bugs make an implicit assumption. Gold is good, therefore it will go up. This is tempting but wrong (ignoring that gold does not go anywhere, it’s the dollar that goes down). One error is in thinking that now you have discovered a truth, everyone else will see it quickly. And there is a subtler error. The error is to think good things must go up. Sometimes they do, but why?   Since putting in a secular low at the turn of the millennium,...
  • The 2017 Incrementum Gold Chart Book
      A Big Reference Chart Collection Our friends at Incrementum have created a special treat for gold aficionados, based on the 2017 “In Gold We Trust Report”. Not everybody has the time to read a 160 page report, even if it would be quite worthwhile to do so. As we always mention when it is published, it is a highly useful reference work, even if one doesn't get around to reading all of it (and selective reading is always possible, aided by the table of contents at the...
  • The Falling Productivity of Debt
      Discounting the Present Value of Future Income Last week, we discussed the ongoing fall of dividend, and especially earnings, yields. This Report is not a stock letter, and we make no stock market predictions. We talk about this phenomenon to make a different point. The discount rate has fallen to a very low level indeed.   We add this chart to provide a slightly different perspective to the discussion that follows below (and the question raised at the end of the article)....
  • Precious Metals Supply and Demand
      Fundamental Developments The prices of the metals shot up last week, by $28 and $0.57.   Heavy metals became pricier last week, but we should point out that the stocks of gold and silver miners barely responded to this rally in the metals, which very often (not always, but a very large percentage of the time) is a sign that the rally is unlikely to continue or hold in the short term. [PT]   Last week, we said:   “One way to think of these moves is...

Support Acting Man

Top10BestPro
j9TJzzN

Austrian Theory and Investment

Archive

350x200

THE GOLD CARTEL: Government Intervention on Gold, the Mega Bubble in Paper and What This Means for Your Future

Realtime Charts

 

Gold in USD:

[Most Recent Quotes from www.kitco.com]

 


 

Gold in EUR:

[Most Recent Quotes from www.kitco.com]

 


 

Silver in USD:

[Most Recent Quotes from www.kitco.com]

 


 

Platinum in USD:

[Most Recent Quotes from www.kitco.com]

 


 

USD - Index:

[Most Recent USD from www.kitco.com]

 

 
Buy Silver Now!
 
Buy Gold Now!
 

Oilprice.com