Objective research to aid investing decisions
Menu
Value Allocations for July 2019 (Final)
Cash TLT LQD SPY
Momentum Allocations for July 2019 (Final)
1st ETF 2nd ETF 3rd ETF

In Search of the Bear?

Posted in Sentiment Indicators

Does intensity of public interest in a “bear market” mean that the bear is already here, as implied by ZeroHedge commentary on Google Trends search intensity? To investigate, we download weekly global Google Trends search intensity data for “bear market” and (for corroboration) “bull market” and relate these data to future weekly S&P 500 Index returns. “Google Trends analyzes a percentage of Google web searches to figure out how many searches were done over a certain period of time. Google Trends adjusts search data to make comparisons between terms easier….To do this, each data point is divided by the total searches of the geography and time range it represents, to compare relative popularity. The resulting numbers are then scaled to a range of 0 to 100.” Using the specified data from the earliest available time on Google Trends (January 2004) through January 2016, we find that:

Because there are many “0” search intensity readings in early data, we start tests with January 2006.

The following chart compares weekly S&P 500 Index level and weekly “bear market” search intensity since January 2006. Visual inspection suggests that search intensity peaks when the stock market is near or at market bottoms.

For precision, we related search intensity to weekly S&P 500 Index return.

SP500-bear-market-search-intensity

The following scatter plot relates next-week S&P 500 Index return to weekly “bear market” search intensity since January 2006. The Pearson correlation between the two series is -0.003 and the R-squared statistic 0.000, indicating that weekly search intensity explains nothing about next-week stock market return.

Could the relationship be predictive at some longer horizon?

SP500-next-week-return-vs-bear-market-search-intensity

The next chart shows correlations between S&P 500 Index weekly return and weekly “bear market” search intensity for various lead-lag relationships, ranging from stock market return leads search intensity by eight weeks (-8) to search intensity leads stock market return by eight weeks (8). For comparison, the chart shows the same relationships for weekly “bull market” search intensity. Results suggest that:

  • Search intensity for “bear market” tends to follow relatively poor stock market performance in recent and current weeks.
  • Search intensity for “bear market” is uninformative about future stock market returns through the next eight weeks.
  • Search intensity for “bull market” also, but more weakly, tends to follow relative poor stock market performance in recent weeks.

In case there is a material non-linear relationship, we look at average next-week S&P 500 Index return by range of “bear market” search intensity.

SP500-weekly-return-search-intensity-leadlag

The final chart summarizes average next-week S&P 500 Index return and average weekly “bear market” search intensity by ranked fifth (quintile) of weekly “bear market” search intensity since January 2006. There are 105 observations per quintile. The average S&P 500 Index weekly return for the entire sample is 0.11%.

There is no progression of stock market performance across quintiles. Nor are extreme quintiles notably informative.

SP500-average-next-week-return-vs-bear-market-search-intensity-quintile

Four other tests show that:

  • The correlation between weekly “bear market” search intensity and weekly S&P 500 implied volatility index (VIX) is 0.54, indicating a material relationship.
  • The correlation between weekly “bear market” search intensity and “bull market” search intensity is 0.31, suggesting that the two searches tend to intensify and weaken together.
  • The relationship between weekly change in, rather than level of, “bear market” search intensity and next-week S&P 500 Index return also has R-squared statistic 0.000. This variable also is uninformative about future stock market returns through the next eight weeks.
  • Restricting the search to the U.S. has small effects on findings.

In summary, evidence from available data does not support belief that public interest in “bear market” per Google Trends predicts short-term U.S. stock market returns.

Cautions regarding findings include:

  • The sample is very short in terms of number of U.S. equity bear markets.
  • Google Trends search data are somewhat unstable from week to week due (at least) to overall series recalculation/renormalization. The cumulative effect of weekly instabilities could be large.
  • Other search terms and search restrictions might affect findings.

See also “Google Trends Predict the Stock Market?” and “Google Search Activity Predicts Stock Market Returns?”.

Why not subscribe to our premium content?
It costs less than a single trading commission. Learn more here.
Daily Email Updates
Login
Research Categories
Recent Research
Popular Posts