Graduate To a Smoother Ride in Your Portfolio

I have to admit that this is my favorite time of year from an investment management perspective. This is not due to the fact that the markets usually enjoy a Santa Claus/Year-End rally. No, this is the time of year that I like to look back on how we did and to plot strategy for the upcoming year.

In looking back, volatility has certainly been an issue that most investors have struggled with this year. There have been many big moves in both directions and more than a couple scary pullbacks (including a -16% correction). The advent of new computer programs has thrust high-frequency-trading into the spotlight on numerous occasions this year with the “flash crash” of May 6th being the poster-child for trading programs and volatility running wild.

The quick moves – especially those that occur in the first (or last) few minutes of the day – are difficult for those of us that do not possess computers designed to react to news and bid/ask inefficiencies within milliseconds. And anyone trying to trade this market has likely been confounded a time or two in 2010.

However, there are things that investors can do to try and smooth out the ride. I recently penned an article on a new style of diversification, Diversification Done Right; A New Take On MPT in which we addressed the concept of diversifying your portfolio by investment strategy. The idea is to incorporate multiple strategies into your portfolio so that you won’t be left high and dry when a particular strategy underperforms or the market environment changes.

For example, in my own personal portfolio, I have mentioned a time of two that I utilize three distinctly different strategies: a market timing strategy (the Daily Decision), a stock management strategy (The Top 5), and a “go anywhere” global ETF strategy (the FlexPro). The goal is to utilize my favorite approaches to portfolio management within my overall accounts. And while it is disappointing when one of the strategies has a bad year, I am experienced enough to know that this is part of the game and the price you pay in order to avoid big mistakes over a long period of time.

This same approach can be applied to managing or trading the stock market. By diversifying your indicators and applying a graduated method of exposure to market risk, one can keep their portfolios in line with the major trends while enjoying a smoother ride along the way.

The game plan here is fairly simple. Instead of relying on one indicator or market model to give you signals of when to be in or out of the stock market, you establish an all-star team of indicators – and then let each everyone on the team “do their thing.”

If you are anything like me, you probably have a handful of favorite indicators that you like to follow. However, the problem is usually the same: How do you know if the indicator you are using is the best for the situation? In other words, when do you follow the indicator’s signal and when do you ignore it? And THIS type of decision is a death knell for most investors.

One solution is to build models of indicators to give you a signal when the “weight of the evidence” tells you the overall health of the stock market is good. The only problem with this is that the all-or-nothing approach to being 100% long or short can expose you to whipsaws.

Pardon the commercial, but our Daily Decision model has been built to get the best of both worlds by staying with the trend and attempting to avoid whipsaws. And frankly, I am pretty darned pleased with our results this year. Sure, we got whipped a couple times and it was tense there for a while, but all in all, the results are respectable.

However, I am never satisfied. So, I keep searching for ways to do things better or to find an approach to add to my personal three-pronged portfolio diversification method. So, this year, I’ve been testing an approach designed to reduce whipsaws even further.

One of the best ways I know of to avoid the pain of getting whipsawed is to use a graduated approach to market exposure instead of an all-or-nothing “timing” approach. While it is true that you won’t make as much money using this method, you also won’t get whipsawed nearly as much.

As I mentioned, I’ve been testing just such an approach for most of 2010 and despite the volatility we’ve seen this year, the results aren’t half bad.

I started by creating a list of my absolute favorite market indicators. But instead of creating a model that would give me a single buy or sell signal, the plan was to have each of the indicators buy and/or sell a set percentage of the portfolio. So, since I am using seven indicators/models, each one controls 14.2857% of the overall portfolio.

Here’s the way it works. If an indicator is positive it contributes 14.2857% long exposure to the overall portfolio. If the indicator is neutral, it contributes 0% exposure. And as you have probably surmised, a negative reading contributes a “short” by adding -14.2857% to the overall exposure level.

So, each weekend, I enter the readings for each of the indicators and simply add ‘em up. In 2010, the exposure levels ranged from -85.71% (seen in late June) to +100%.

Intellectually, I feared that since there was no leverage and it is tough to get the model to be 100% short, that I would wind up underperforming the market. I also figured that since the model was weekly, it might miss some of the big blasts. And while both situations have indeed occurred, the benefit of staying in tune with the overall market outweighed the negatives. For example, as of December 3rd, the system shows a gain of +20.12%, which is more than double the 9.83% of the S&P 500.

Granted, this return is not as good as our weekly “timing” model, which is up nearly 28% on the year. And in all fairness, using this approach on a daily basis would be smarter. But, despite the caveats, the volatility was lower and the model account for the graduated approach has been positive each and every week of the year – so the ride has indeed been smoother.

My guess is that there are probably a few readers that might be interested in testing this on their own. So, here’s a list of the seven indicators/models I am using. It is safe to say that this isn’t a perfect list – but it’s my list nonetheless:

• 15 Month Weighted MA (moved forward 2 months)
  
• 9 Week Weighted MA (moved forward 2 weeks)
  
• The Weekly Daily Decision Model
  
• The Short-Term Trend component of the Daily Decision Model
  
• The Intermediate-Term Trend & Breadth Confirm component of the Daily Decision Model
  
• The Intermediate-Term Volume Thrust component of the Daily Decision Momentum Model
  
• “Manager Discretion