Full Trend Analysis of ≈750 ETFs and ≈1000 Stocks.

Philosophy and Approach

(We crunch mountains of data into nuggets of investment insight)


Our Philosophy

Our philosophy is to be systematic and disciplined, and focus on just a handful of key action items, because research shows we can only focus on four major variables at one time. We mostly use an approach that is trend-following (or momentum-based) in nature, combining different time-frames and methods of analysis. For example, we like to confirm price-based or “technical” with “fundamental” data on the US economy.

A Sports Analogy (Or, what happens to golf scores when the wind picks up?)

Unlike professional golf, where you can practice for a few days on a golf course before the contest begins, where the hazards don’t move, and you can fine-tune your shots, investing affords no such opportunity to prepare or practice. World-class athletes work endlessly to optimize their equipment and themselves, because they live in a universe of cause-and-effect. But note what happens to their performance when playing conditions are not ideal, i.e., a key assumption is violated. Observe what happens to golf scores when the wind picks up, or to lap-times in Formula One when it rains: they both go up. Now imagine having to compete at a world-class level without preparation, practice or optimization of any sort. But that is precisely what the markets demand, because markets operate in a world full of randomness, or a world of diverse causes, and variable, and imprecise effects. Since market conditions are rarely ideal, often confusing, and constantly changing, our robust tools can help you navigate this difficult terrain.

A buy-and-hold strategy on your part may be good for the money manager who gets paid for total dollars (assets) under management. But is it the best proposition for you? You don’t have to just buy-and-hold and “diversify” your portfolio with products that become highly correlated in a crisis. There is a lesson from sports we can apply to investing: defense wins championships. Our models can potentially help you on defense (i.e. 2008), when “cash is king”.

The Paralysis of Analysis (Torrents of Data, Too Many Tools)

Every investor, money manager, investment advisor or trader faces an enormous flow of often contradictory and confusing information, which must be converted into a few usable, actionable items, so you can do something about it, here and now. A fundamental analyst must deal with lags in data, fragmentary or incomplete data sets, geopolitical gyrations, central-bank shenanigans, shifting correlations, and changes in underlying technology, which call into question using decades of data. A technical analyst must choose among dozens of indicators, which have some degree of correlation, but which suggest different actions, especially when analyzed using different time-frames. Data, data everywhere, but not a drop of insight!

Simplify, Simplify, Simplify: We boil-down the data (so you don’t have to)

The most difficult challenge is to simplify the entire investment process. The easy thing to do is to give you a toolkit of a hundred indicators and massive data sets. The laws of thermodynamics prove that reducing entropy takes work. There are a million different ways to combine technical and fundamental data into trading systems or decision systems, each with its own strengths and weaknesses. We have done the hard work, of choosing one particular set of decision variables, from millions of possibilities, based on original research, as well as years of tough trading experiences in the double-edged world of the futures markets.

An Objective Process (Free of subjective embellishments)

Our approach is derived from decades of experience developing trading systems for the futures markets [1, 2], quantitative models [3, 4] and novel indicators for technical analysis [5..8]. So, we have resolved the challenges of considering different time-frames of analysis, and different fundamental and technical measures of price action. We have found a combination we like. It is not perfect, and will produce losing signals, but, if applied consistently, in a disciplined way, will generally stay on the right side of the major trend, using objective analysis, free of subjective embellishments.

So, what happened in 2008? (The models turned bearish quite early)

Our indicators based on fundamental data on the US Economy from the US Federal Reserve turned bearish in early 2008, months before the collapse of Lehman Brothers (see Figure 1). Our Strategic Portfolio Allocation using the S&P-500 data turned mostly to cash well before the October declines (see Figure 2). For the record, these calculations are using the current constituents for each index, but the process is mechanical and would likely have produced essentially the same results in real-time. Note that the Strategic Portfolio View tends to use a relatively long time period for its calculations.

And, what about 2009? (Strategic models turned bullish mid-year)

Just as the models turned bearish in 2008, they responded to the Federal Reserve easing by turning substantially bullish by the middle of 2009 (see Figures 2). Now, this bullish turn occurred at a time of great fear and uncertainty, when gloom-and-doom was the preferred potion of the day. So, our models did their job. Would you have followed suit?

A word about the “speed” of model response (Models don’t work every time)

One of the key challenges in designing trading systems is to calibrate the speed of the response to changes in market direction. Due to automation and computerization, markets can change direction rapidly. A model that responds rapidly would require rapid portfolio changes as well, and could produces a few false trades as the market searches for direction. A model that “smooths” or averages the data may seem to respond quite slowly at key turning points. The correct speed for any moment is not always obvious. Our Tactical models respond quite quickly, whereas the Strategic models are slower to turn. However, our models are not designed to respond to “flash crash” scenarios.

There is no “right” answer, but simplicity is worth something

Ultimately, there is no trading system that works perfectly every time. The key advantage we seek to provide is the ability to convert a large amount of data into a decision dashboard. We hope to save you time and simplify a small part of your life. Not a bad deal at all in today’s world of twitter trends leading to bitter ends.

Selected References

  1. “The New Technical Trader”, Tushar S. Chande and Stanley Kroll, Wiley [1994] ISBN: 0-471-59780-5
  2. “Beyond Technical Analysis”, 2nd edition, Tushar S. Chande, Wiley [2001] ISBN: 0-471-41567-7
  3. Tushar Chande, “Rho Trend Barometer”, The Hedge Fund Journal, November 2011
  4. Tushar Chande , “Price Range Compression Explains Recent CTA Returns,”, The Hedge Fund Journal, August, 2014
  5. Tushar Chande, “Estimating the depth and duration of drawdowns from past performance data”, The MFA Reporter, September, 1998
  6. Tushar Chande, “Controlling risk and managing investor expectations by modeling the dynamics of losses in hedge funds and alternate investment strategies”, Derivatives Quarterly, 5(3): 52-58, Spring 199
  7. Tushar Chande, “Adapting moving averages to market volatility”, Technical Analysis of Stocks and Commodities, 10(3), 1992.
  8. Tushar Chande and Stanley Kroll, “Stochastic RSI and Dynamic Momentum Index”, Technical Analysis of Stocks and Commodities, 11(5), 1993


Figure1: The Coincident Indicators Rating, a “fundamental” measure, converts the coincident measurements of US economic activity into a scale from 0 to 100, where readings above 70 are bullish (expansionary) and readings below 30 are bearish (recessionary). Despite data lags, this indicator dropped into bearish territory, and then to zero in early 2008, well before the rapid declines in the fourth quarter of 2008. We expect stocks to decline with or before the coincident measures, in anticipation of prolonged economic weakness.



Figure 2: We show the “technical” Strategic Model allocations at each month-end starting January, 2007 through December, 2009 for the large-capitalization S&P-500 index. The Strategic Model responds relatively slowly to market action, and switches between an S&P500 ETF (say SPY) and cash. Its binary allocation can be 100% in the market, or 100% in cash. By the end of February 2008, the model was about 43% to market, and 57% in cash, a defensive position maintained throughout 2008. By the end of September 2008 the model was approximately 67% in cash. By mid-2009, the relative allocation began leaving cash behind. By the end of June, 2009 it was 62% in the market, only 38% in cash. By August, 2009 this model was holding only 7% or so in cash, i.e., it was essentially fully long, at a time the prevailing mood was full of doom and gloom.