Want to keep tabs on investor sentiment, but don't have a lot of time?
We created a handful of models, with time frames between intraday and weekly, that aggregate many of our sentiment indicators. You can see in one quick glance where sentiment is at any given time. We also provide historical data for reference and testing purposes.
These are by no means a market-timing panacea, but they each serve a useful purpose when put into the proper perspective.
We want to provide users with the information they need to be aware of the current state of market sentiment, in all time frames. Like fundamental and technical analysis, sentiment can be quantified, measured and tested.
That's what we have done to come up with our models.
Time Frame: Very Short-Term
The STEM.MR model has the shortest time frame.
It was developed because several of the shorter-term sentiment indicators we follow have displayed a mean-reverting tendency in the past - when they go too far in one direction, they tend to snap back to an average value.
Often when they snap back, it correlates with short-term market turning points. We use 30-minute data to calculate the index.
Depending on market conditions, a signal could last anywhere from an hour to several days. Typically, however, the market makes a decision one way or the other and it is usually clear soon whether the signal will be effective or not. These extremes typically happen three to four times per month.
Time Frame: Short-Term
The Short-Term Extreme Model (STEM) is a model consisting of several sentiment measures including:
- Put/Call ratios
- Proprietary sentiment measures
We use 30-minute data to calculate the index, but it covers a longer time period than our shortest-term model.
An extreme model reading can be considered anything outside of the model's trading bands. A reading outside the upper band would suggest that the market is overbought and a reading under the lower band would suggest that it is oversold.
Time Frame: Intermediate to Long-term
We track a large number of indicators on a daily and weekly basis that can be separated into "dumb money" and "smart money".
The dumb money indicators are typically made up of retail traders and trend-followers. This is NOT to say that all (or even most) retail mom-and-pop investors, and certainly not most trend-followers, are "dumb". In fact, they are by definition correct during the bulk of a trend.
It is only when these traders move overwhelmingly into a bullish or bearish position on a market that their timing is suspect. This is often when a trend is susceptible to a reversal. When everyone moves to one side of a boat, that boat is likely to tip.
The smart money indicators are mostly made up of institutional accounts. These traders are often hedging day-to-day moves in the market, and therefore are often trading against the prevailing trend. Again, it is only when these traders move to an extreme that a market is most likely to reverse in their direction.
The spread between the two groups of traders is an effective way to see how they are positioned against each other. When the dumb money is very long, and the smart money is very short, then markets tend to struggle going forward, and vice-versa.
Time Frame: Intermediate to Long-term
The Advisor & Investor Model (AIM) is a model which averages the momentum of the four major sentiment surveys.
This model takes advantage of the fact that when the typical investor and investment advisor should be most bullish, they are most bearish. And, when the markets are getting overbought and are about to turn, these Johnny-come-latelys are most bullish.
When optimism reaches a degree extreme enough to move the model above its upper trading band, then we know that investor and advisor sentiment is reaching a frothy stage that is very rarely rewarded with rising market prices - usually, the market stages a decline soon afterwards (red arrows highlight those times on the chart below). The arrows do not depict buy and sell signals, but rather times when the model reached far outside its trading bands and then reversed.
In contrast, when the survey populations are so pessimistic that it forces the model below its lower trading band, then we know that we have likely entered a panic selling stage that will result in a market rebound with a high degree of accuracy (the green arrows on the chart).