Many experts feel that mixing fundamental and technical analysis is like trying to mix oil and water, but I believe that screening fundamental data and viewing weekly chart patterns are equally important in measuring a stock's risk/reward.

The difficulty that comes with interpreting both data sets is that when all factors are equal, a stock becomes more fundamentally positive when its price goes down and less fundamentally positive as the price goes up. But when the price of a stock is moving higher, the technicals tend to improve, and when the price of a stock is declining, the technicals tend to deteriorate. This makes combining the two disciplines extremely difficult, but in my opinion, it is a necessary challenge.

The way I go about balancing the two is to view the economy and the markets as a system, and look for cause-and-effect relationships among the variables that can lead to trading decisions in the U.S. capital markets, and in stock evaluation and selection.

That is part of the reason I use the ValuEngine Stock Valuation Model in my fundamental valuations. The model supports my theme of technology leadership for 2005, as technology began 2005 undervalued by 11.78%. The table at the bottom of the page cleanly lays out the case relative to other sectors.

This model employs a three-factor approach to stock valuation: a company's trailing 12-month earnings per share, the analyst consensus estimate of the company's forecast 12-month EPS, and the 30-year Treasury yield. Those variables are all combined and used to create a more accurate reflection of a company's fair value. Armed with these framework features, the model paints a detailed picture of a company's fair value, which is represented by the "model price." Below are some of the variables that lead to the model's price of a stock and calculation of the percentage undervalued or overvalued.

Firm-specific variables:

Correlation between the company's earnings per share and the interest rate environment.

Long-term earnings growth rate.

The duration of the company's business growth cycle.

The company's systematic or beta risk.

Interest-rate-related criteria:

Long-term history of interest rate climate.

Interest rate volatility.

The duration of the interest rate cycle.

Short- and long-term historic factors in the model's calculation include past valuation levels of the stock and its recent price momentum factor relative to other stocks. These considerations applied with the firm-specific variables allow the model to differentiate a stock across sectors and within the company's own business growth stages.

The model price uses 12-month historic and forecast EPS values and the current 30-year Treasury yield as primary determinants. In calculating risk/return values such as the Sharpe ratio, the historic or forecast EPS periods are five years. Overly simplified, here's the effect of these factors on the model price: When the yield on the 30-year Treasury bond is declining, the model price of the stock increases. When the yield on the 30-year Treasury is rising, the model price decreases. When the current or projected earnings per share increases, the model price increases.

Technical Analysis Tools

In 1984 I developed proprietary forecasting models that help me evaluate the risk/reward for markets over a rolling six-month horizon. My models are based upon a matrix of the past nine closes in these time horizons -- daily, weekly, monthly, quarterly, semiannual and annual. In my judgment, the evaluation of nine years of closes builds in the summation of all bullish and bearish events, which is necessary to project future volatility including supports, pivots and resistances. Bearish fundamentals may cause market weakness that will test support. Bullish fundamentals may cause strength that will test resistance. Pivots tend to act as magnets that have an 85% chance of being tested during their time horizon.

My proprietary analytics assume that nine years of closes are enough to incorporate all possible bullish and bearish events that cause volatility in any market.

The Dec. 31, 2004, close established new inputs to my monthly, quarterly, semiannual and annual models, which helps me determine what I project as the risk/reward over a rolling six-month horizon.

The standard tools of technical analysis, or TA, that I use were established in the fourth quarter of 1987.

My goal was to establish a weekly measure of positive, neutral or negative technicals with a method that resulted in the fewest false signals. Following some "brute force" back-testing for bond futures, I decided to use a weekly 12x3 slow stochastic reading for momentum, and the five-week modified moving average as the trigger, given a weekly close above or below that level.

The 12x3 weekly slow stochastic

reading is between 00.0 and 100.0, where a rising trend above 20.0 is positive and a declining reading below 80.0 is negative. Under 20.0 is oversold. Above 80.0 is overbought.

The five-week modified moving average

starts with a five-week simple moving average. The five-week MMA is then calculated as

(Prior five-week MMA * 4) + (Latest Close) / 5.

The 200-week simple moving average

is considered a support or resistance and an indication of the longer-term trend.

As part of the interactive portion of Technology Report

, I will include alerts as key technical statistics are tested, violated or achieved. There will also be alerts based upon ValuEngine data points as the price action or new information from data sources moves a stock through VE overvalued or undervalued thresholds.

Signals: VE/TA Buy, Sell or Hold

VE buy:

40% or more undervalued. A stock needs to be at least 40% undervalued according to the ValuEngine Stock Valuation Model to be considered a candidate for a long-term portfolio.

VE hold:

Between 40% overvalued and 40% undervalued.

VE sell:

40% or more overvalued.

TA buy:

Weekly close above the five-week modified moving average with rising or overbought 12x3 weekly slow stochastic.

TA hold:

Neither a TA buy nor a TA sell.

TA sell:

Weekly close below the five-week modified moving average with declining or oversold 12x3 weekly slow stochastic.

Richard Suttmeier is president of Global Market Consultants, Ltd., and chief market strategist for Joseph Stevens & Co., a full service brokerage firm located in Lower Manhattan. At the time of publication, he had no positions in any of the securities mentioned in this column, but holdings can change at any time. Early in his career, Suttmeier became the first U.S. Treasury Bond Trader at Bache. He later began the government bond division at L. F. Rothschild. Suttmeier went on to form Global Market Consultants as an independent third-party research provider, producing reports covering the technicals of the U.S. capital markets. He also has been U.S. Treasury Strategist for Smith Barney and chief financial strategist for William R. Hough. Suttmeier holds a bachelor's degree from the Georgia Institute of Technology and a master's degree from Polytechnic University. Under no circumstances does the information in this commentary represent a recommendation to buy or sell stocks. While he cannot provide investment advice or recommendations, he invites you to send your feedback to