Tips on Technicals - RSI vs. Stochastics
Momentum indicators have proven themselves to be useful indicators for forecasting market trends and changes. Those that are indexed also enable the analysis to be applied to all markets with equal ease and consistency. However, not all oscillators are equally valid in all types of markets. This issue of "Tips" will compare and contrast two of the most widely used oscillators, the Relative Strength Index and Stochastics, and examine the conditions under which each is most reliable.
RSI
The Relative Strength Index (RSI) was developed by J. Welles Wilder, Jr. As a momentum oscillator, the RSI measures the velocity of price movements. In this model prices are generally considered to be elastic in that they can move only so far from a mean price before reacting or retracing. Rapid price advances result in overbought situations and rapid price declines result in oversold situations. The slope and values of the RSI are directly proportional to the velocity and magnitude of the price move and are extremely helpful in identifying overbought and oversold situations.
The core of the formula for RSI takes the last "n" periods and divides the gross positive changes per period by the gross negative changes. This means that the more often prices move higher in that "n" period span and the greater those changes become, the higher the RSI value. By depending on both the number of up closes and the magnitude of those closes, RSI filters out normal volatility difference between markets while maintaining the significance of single large price moves. By reducing the number of periods in the calculation, RSI can be made more sensitive (faster).
The RSI value itself ranges from 0 to 100 and support, resistance and market trends can be found on the RSI curve. Generally speaking, an RSI value above 75 indicates a possible overbought situation and a value below 25 indicates a possible oversold situation. This does not mean, however, that a market will immediately reverse once either of these levels is reached. It is more likely that the market will pause to consolidate, resulting in a more neutral RSI value. The RSI chart is most often used in conjunction with a bar chart. Relatively high RSIs (55-75) normally accompany a positive price trend and relatively low RSIs (25-45) normally accompany a negative price trend. Divergences between price action and the RSI plot could signal market reversals.
Stochastics
The Stochastic indicator, developed by George Lane, can be a valuable tool for identifying near term tops and bottoms to help in timing trades closer to local reversal points. A Stochastic is the measurement of the placement of a current price within a recent trading range. The theory is that as prices rise, closes tend to occur nearer to the high end of their recent range. When prices trend higher and closes begin to sag within the range it signals internal market weakness.
Stochastics is one of a few indicators that uses two lines, known as the K and D lines (sometimes known as %K and %D). The D line is simply a smoothed version of the K and the two of them are analyzed for both overbought and oversold situations just like RSI. Two lines give the added dimension of crossovers, which are similar to price crossovers of moving averages.
Comparisons and Contrasts
Simply stated, the Relative Strength Index yields the most meaningful results in trending markets while Stochastics work best in flat or choppy markets. While the goal of each is similar, they were designed with different specific purposes. The RSI, as mentioned, helps determine when a price has moved too far too fast. This implies a trending market. Stochastics help determine when a price has moved to the top or bottom of a trading range, which implies a non-trending (flat or choppy) market.
In an earlier edition of "Tips" we discussed how momentum indicators like RSI and Stochastics were used to find possible market turning points ("Divergence"). We also covered how to apply trend and support/resistance analysis to the indicators themselves. Here, we will limit the discussion to when to apply these two indicators, rather than how to read them.
Figure 1
Figure 1 shows 100 days of daily trading for December 1994 Comex Silver with 9 day RSI and Slow Stochastics. Note how RSIs remained between 40 and 25 for most of the fourth quarter decline, never reaching the neutral 50 mark. Stochastics oscillated from a bullish 60 to an oversold 5 giving false indications of a market reversal. RSI was effective because the instrument is trending.
Figure 2
Figure 2 shows 14 days of hourly trading for Time-Warner Inc. with 14 day RSI and Slow Stochastics. This stock traded in a half point range for three days. RSI values converged around the 50 mark indicating neither strength nor weakness. Stochastics, while rising for over two days, remained mostly in the bearish range under 40 and exhibited a bearish crossover early on the third day. RSIs did not indicate the trading range been broken down until it actually happened. For a non-trending instrument, Stochastics proves more reliable.
While both RSI and Stochastics are useful in most markets, each has its own specialty. As with all technical indicators, multiple indicators give more valid signals than individual indicators.
