The Investopedia Anxiety Index (IAI) is a measurement of investor sentiment based on the behavior of 30 million Investopedia readers around the world. We track reader interest in 12 financial terms and measure the percentage increase or decrease in those searches over time.

Current IAI Reading

For a current reading of the IAI, see our Index landing page.

The IAI measures reader interest in topics from investing to personal wealth, while the VIX — formally known as the CBOE Volatility Index — is a popular measure of the stock market's expectation of volatility implied by S&P 500 index.

We track the Anxiety Index against the VIX to see if interest in the terms we measure precedes movements in the VIX. In fact, it does, which demonstrates that our readers typically come to learn more about a topic before they take action with their investments.

Some Background

In 2012, Seth Steven-Davidowitz published an article in the New York Times explaining how he used Google search results to uncover voter bias that pollsters were unable to find. Investopedia has 20 million monthly unique visitors, and with Steven-Davidowitz's work in mind, we asked ourselves, “What can the search behavior of our readers tell us about the state of markets and the economy?"

We have the data: more than 100,000 URLs of quality content going back before the collapse of Lehman Brothers and the 2008 financial crisis. I represented the editorial team and partnered with Ronnie Jansson in our data science division at the end of 2015 to search for patterns in our most highly trafficked materials. We carefully selected a dozen terms on topics that suggested investor fear, like "default," and opportunistic terms, like "short-selling."

Finding a signal in noisy web traffic data is difficult due the varied seasonality of our readership (for instance, traffic declines on the weekends) and exogenous factors like Search Engine Results Page (SERP) rank. We first needed to develop a methodology to remove this noise and produce an index that robustly tracks the actual ebb and flow of interest in the chosen topics. The result is the plot below.

When we looked at the results of the analysis the first time, we found that the major peaks in the index occurred exactly where they would make sense: around major events like the fall of Lehman Brothers (by far the most significant peak), the Greek debt crisis and the U.S. credit downgrade by Standard and Poor’s.

We used 13 terms in the very first version of the index. Many of those, however, had few page views and contributed little to the index. In the final version of the IAI we used 12 urls, all with high page view counts. We also now use several thousand more term pages in the normalization procedure. In total we used close to one billion page views to produce the nine-year monthly IAI plot.

We had set out to create a proxy or index for investor sentiment, but we needed an outside point of reference. The Chicago Board of Options Exchange’s Volatility Index (VIX), often referred to as "the fear index," is commonly used as a gauge of investor fear. We plotted the VIX next to our new creation, and the results spoke for themselves:

Over a period of almost a decade, the large scale features are very similar in the VIX and the IAI despite measuring different phenomena (stock market volatility and content consumption, respectively). It gets even more interesting when the two are overlaid on top of one another:

Perhaps the most compelling comparison is at the very earliest point of the plot. For more than a year prior to the peak of the financial crisis in September 2008, the IAI was profoundly elevated (around 120 or so – a level that had not occurred in a single month in the most recent four years), while the VIX remained subdued, around 20. In other words, based on the VIX alone you would be caught completely off guard by the biggest financial crisis of our generation, whereas the IAI was an alarm blaring for more than a year before the crisis hit.

Since the Anxiety Index measures interest in terms that include both market related like Correction or Bear Market, as well as personal finance terms like bankruptcy and debt, its possible and not uncommon for the overall index to show 'Low Anxiety', while the Market Indicators may show 'High Anxiety'.

That's exactly what we are seeing as 2018 comes to an end. The stock market entered into a correction in November and finally into a bear market on Dec. 24. Anxiety levels are high for the Macroeconomic and Markets based terms, but low for Personal Finance terms.