Trading Strategy – A Guide to Market Correlations

Trading Strategy – A Guide to Market Correlations

Financial markets rarely trade completely in a vacuum. What happens in one market frequently spills over and may drive other markets’ movements. For traders using predictive market software like VantagePoint, the intermarket relationships are analyzed, evaluated and weighted to forecast price trends.

Still, it’s important to understand the theory behind intermarket relationships as they apply to complex market conditions. Because what you may have learned in your college economics course may not apply to trading today’s correlated markets.

Correlation Basics

Without getting into a major statistics lesson, a brief look at the basics of correlations is a good idea. First, a coefficient of correlation is a number between +1 and -1 and represents the degree of statistical interconnection between two assets. A correlation of +1 indicates a perfectly positive correlation and means that two assets’ prices move in perfect tandem. A correlation of -1 expresses a perfectly negative correlation and shows that the two assets’ prices move perfectly inversely to each other. A correlation of zero means the two assets’ price moves are statistically uncorrelated, meaning the price movements in one asset have no statistical meaning for the other asset’s price direction. Correlations between +0.3 and -0.3 are typically viewed as weakly and unreliably correlated; coefficients of +/- 0.5 are considered significant, while coefficients of +/- 0.7 are considered quite strong, statistically speaking.

Correlations are calculated based on a number of observations or periods, such as weekly, daily or hourly closing prices. As such, one of the most important elements to keep in mind when interpreting the significance of any correlation is the number of observations used in the calculation. A correlation calculated using only 20 or 30 periods is likely to be of minimal reliability, but coefficients relying on 100 or more observations typically have a high level of statistical significance and reliability.

Why Do Correlations Exist?

The longer answer is that correlations exist as a function of time and, most important, a series of fundamental economic relationships becoming elevated for a period. The key with the time component is that just as time passes, correlations come and go, lasting anywhere from weeks to months or years. And they constantly vary in strength. The point of the fundamental economic relationships portion is that there is nearly always a real world relationship between assets that account for much of a correlation’s existence. Those relationships are in turn driven by changing economic and market environments that adjust over time. Another important reason behind the frequent appearance of intermarket correlations is you! The traders who actively speculate, hedge and invest based on correlations, whether intuitively or programmatically, generate a self-fulfilling feedback loop in the process.

The downside here is that when correlations break down, many traders are on the same side, exacerbating the resulting divergence and market fallout. That’s why it’s important that your trading software not only uses intermarket analysis, but also employs predictive indicators to forecast trends.

Short Term Trading and Market Correlation

Most statistically significant correlation studies span long time horizons. Unfortunately, they provide little reliable insight into shorter-term price relationships, which is where most traders focus. The result is that ostensibly time-tested relationships can, and frequently do, break down in the short run (intraday or during a few days). As a result, traders employing correlation based trading strategies need to remain especially alert to short-term divergences in addition to longer-term shifts in the underlying economic environment.

Again, enter the importance of software like VantagePoint. Relying on its proprietary, patented technologies that apply neural network pattern recognition to intermarket data and its patent-pending technologies that then create leading indicators, VantagePoint helps spot the short term shifts in market correlations.

Still it’s important to remember that correlation is not causation. Just because two assets show a degree of correlation does not mean that movement in one is necessarily causing the other to change. Especially on an intraday basis, it is critical to have a sense of which market is leading and which is following. The primary catalyst will usually be some piece of fundamental news or data, but you need to find that out and discern which market is dominating at any given moment. Technical analysis of correlated markets is also required, as the lagging asset may suddenly play catch-up if it breaks an individual technical level.

This requires that your technical indicators to be leading and not lagging!

Correlations are irrelevant if your Technical Analysis Can’t Keep Pace

Many technical indicators, such as moving averages, attempt to filter out short-term price fluctuations so that the underlying trend can be observed. A side effect of doing this is that the technical indicators tend to lag behind the market. Such technical indicators are referred to as lagging indicators. This lag effect typically causes the trader to respond late to market changes, resulting in lost profit opportunity and risk of increased losses.

Still, moving averages are very popular because they smooth out the movement in prices, are easy to calculate and understand, and depict the underlying trend.  But, the lagging nature of moving averages has always been the bogyman that has kept them from realizing their true potential.

VantagePoint employs proprietary computer processes which address these limitations and overcome the lag effect through the development of methods, systems, and devices that combine both actual and predicted data derived from the application of neural networks to intermarket data found to be most influential on each specific primary market.

In one aspect of the invention, an algorithm is used to integrate the predicted data with actual technical indicator values to create a hybrid technical indicator that overcomes the lag effect that was previously thought to be an inherent aspect of using technical indicators.

Again, market correlations don’t move markets, a confluence of factors do and this requires state of the art, predictive technical analysis for traders looking to capture these relationships.

VantagePoint Software Forecast

Related Articles

Go to Top