The Zettacap Blog

Zettacap allows investors to leverage the power of big data.

Learn More

EBAY, when negative Sentiment turns into a bullish breakout

Published: March 4, 2014

By: Kevin Coogan

Lesson learned: use sentiment but keep your eyes on other variables to understand what is happening in the stock and watch for non-confirming price action.

A major trend in leveraging Big Data for investing is to focus on NLP polarity analysis of electronic communication, more specifically social media like twitter (cashtags). The idea is straightforward – score tweets, posts, blogs, news, and/or other communication for positive and negative sentiment. A significant spike in positive sentiment is a buy signal, whereas a spike in negative sentiment is a sell signal. Such signals receive further reinforcement when accompanied by spikes in volume of mentions.

For extremely short-term trading lasting from a few seconds to a few days, such an approach can make sense. An assumption of course is that you are able to take advantage of such shifts in sentiment quick enough. For traders and investors looking at slightly longer term holding periods, the sentiment indicator becomes yet another variable that should be incorporated into a more holistic analysis which includes looking at price action or the movement of the traded stock price.

As an example, we can look at what happens near the end of bear markets. They do not end on the day that horrific news is replaced by glowingly positive news. Generally they end when prices stop falling on negative news and even move higher in the face of continued negative news (albeit slightly less negative news on the margin, which is the topic for a different post). At such a stage, the markets implicitly state that they have already incorporated contingencies for such negative news and that they likely overshot a bit in terms of valuation. The actual positive news begins well after the actual market low. In other words, if you were trading on simple binary NLP polarity sentiment under these conditions you would not have done very well by waiting for news to become positive. In this case, the key would have been to follow sentiment and price action and look for divergences which imply an important change in market direction.

The same thing occurs in specific stocks. Many times stocks will hit a particular level and resist flashes of sentiment spikes. In other words, you could have a directional movement due to the sentiment shift but the price would return rapidly or even begin to move in the opposite direction. Astute investors realize that when stock prices become resistant there is something else going on that requires a closer look.

The point is that beyond very short-term trading scenarios, sentiment variables should be taken in context with other variables such as stock price action and with the stage of the stock in general.

The case of EBAY is very interesting in that it shows how a negative sentiment spike helped the stock breakout of a yearlong consolidation. This is somewhat counterintuitive as you would expect a large negative sentiment spike to result in a significant and sustained decline – in EBAY’s case, the opposite occurred in that it ended up breaking out of (going up through) a range that had trapped it for quite a while.

We will look at a number of images and take into consideration, stock price patterns, spikes in mentions and sentiment, short selling volume, and an eventual breakout. This study basically consolidates traditional technical analysis of stock trends, sentiment analysis, and short selling analysis – thereby marrying popular technical price trend analysis with sentiment analysis.

First, let’s start by looking at the stock chart for EBAY.

EBAY stock price

EBAY stock price

It is apparent that the stock has been basically moving sideways for about the last year. It can also be seen that a similar trend occurred in 2011. These trends can easily be highlighted into channels or flags as is shown in the following image.


EBAY stock channel consolidations

EBAY stock channel consolidations

Channels, similar to the ones in the previous image, occur fairly regularly and are a well-known phenomenon. Trend followers especially like such formations as moving out of a channel in either direction normally signifies a rapid movement. A breakout is referred to as moving upwards past the upper portion of the channel and a breakdown refers to falling below the lower portion. The problem is that it is difficult to know if it will break to the upside or downside, or when such a break will occur.

It looked like the timing for a breakout or breakdown became clearer in late January as the company released its quarterly earnings and Carl Icahn began to push for the company to sell its Paypal unit. Predictably, this generated a considerable amount of attention and mentions, as seen in the following image.

EBAY stock price vs mention volume

EBAY stock price vs mention volume

The spike in mention volume dwarfed what had come before it. This was a seismic event according to the attention it received. Because the stock price was fairly close to the upper portion of the range, you would expect a breakout if the interpretation was solidly positive. The aggregate sentiment was negative on the day – and it soon fell rather sharply. Judging from the fact that it had failed to breakout on such sizeable news, a trader would likely expect the stock to move to its lower portion of the channel and test that level.

The implied negative sentiment is confirmed by the spike in short selling volume that occurred on the same day, as is seen in the following image.

EBAY short selling volume

EBAY short selling volume

The corresponding spike in short selling volume was the highest for EBAY during the sample period. It confirmed the general negative interpretation.

Although the stock price declined for the following few days, its decline was in fact rather muted considering the measurement of mentions and short selling. When the stock began to appreciate after a moderate decline, this was a signal that perhaps after going sideways for about a year, all of the potentially bad contingencies had been incorporated.

The next major move for the stock was in fact a breakout (to the upside), which can be seen here.

EBAY stock channel breakout

EBAY stock channel breakout

This breakout was forceful with the stock moving quickly beyond a trading range that had plagued it since the beginning of 2013. It occurring within a few weeks of record-breaking highs in mention volume, in total negative sentiment, and in short-selling volume is a testament to the importance of following multiple variables to understand the situation and stage of a stock and not to simply focus on one variable.

1. The stock declining from January 23 to January 29 gave sellers ample opportunity to benefit from leveraging the negative NLP polarity sentiment produced by the after-hours January 22 announcements. So, multi-day sentiment traders actually could have profited in this window.
2. There was no further follow through, even with the weight of the short selling behind it.
3. The fact that the stock began to appreciate towards the upper limit of the channel again, so soon after the spikes in mentions, negative total sentiment and short selling should have been a major flag for investors. In other words, if a bearish burst of activity was not enough to send the stock significantly lower the danger of an unexpected upside movement was fairly high.
4. The breakout from a yearlong consolidation soon after these events depicts how markets can actually move in the opposite direction of a shock once they determine that there will be little follow through.

Sentiment is Dead, Long Live Sentiment

Published: February 28, 2014

By: Kevin Coogan

Sentiment trading has been around a long, long time – and will continue to be. In this post, we argue that, although extremely important and integral to trading, the market’s focus on polarity sentiment for short-term trading is a bit exaggerated and that other uses and new forms of sentiment will begin to augment it.

Sentiment directly impacts asset prices through somewhat predictable shifts and overshoots – anyone who has spent a considerable amount of time on a trading desk can attest. Though sentiment has been around, its quantification is rather ‘new’, and its quantification in real-time even more so.

Measurement began with polls and surveys. These were and are still great. The problem of course is that they lack a real-time element. Additionally, issues with respondent selection are apparent.

Social media and the new world of full transparency have opened up financial market analysis to more modern types of sentiment. Looking at tweets, cashtags, other micro-blog posts, news, and/or user generated content can greatly improve insights into sentiment.

Why wait for an end-of-day or even end-of-week survey to be released? Sentiment can be yours right now. It is quicker and can be measured with an impressive amount of scientific certitude. All of this makes perfect sense and is a large benefit of the advent of Big Data.

The issue that has appeared, however, is that positive / negative polarity natural language processing (“polarity NLP”) sentiment has become too central when discussing Big Data and finance. In fact, when you mention Big Data and investing to most people, they automatically think of two things:

  1. sentiment based on polarity NLP scoring of social media, and
  2. short-term directional trading based on same.

Yes, these are important, but if your horizon ends there, you are frankly missing so much that the next generation of financial data and analysis has to offer. Furthermore, the rather clear impression is that going forward you will miss even more as Big Data continues to produce revolutionary insights.

Let’s state that sentiment derived from NLP polarity analysis is extremely important. It offers rapid interpretation of events and of market moods. It can also provide very interesting insights into market trends that would not be apparent otherwise. This post is not arguing to not use this type of sentiment, just that the market should better understand its limitations and be open to other new forms of data and even other measurements of sentiment.

To help portray these points, we will publish over the coming days two posts.

The first highlights a somewhat counter-intuitive use case for negative NLP polarity sentiment. To this point, the generally accepted interpretation is that an unusually large negative sentiment spike offers a great selling or short selling opportunity with potential follow through. In this case, the reverse is shown to be true in that after a few days of declines the stock bolted higher – thereby reinforcing the 360 degree approach that is required when interpreting any single variable.  In other words, do not rely or focus on just one variable.

The second highlights a new form of sentiment and how it can be used. Specifically, it highlights (what we believe) is a first in directional sentiment analysis, which has proven both intuitive to understand and highly useful in its application.

Big Data and Finance, Speed versus Insight

Published: February 26, 2014

By: Kevin Coogan

Big Data will revolutionize finance and investing. Nearly all investors will, at some point, begin to take advantage of the benefits it provides. Some, however, will leverage Big Data to focus on the speed at which information is gathered and understood, while others will prefer focus on deep comprehension and new insight based on this tidal wave of data.

The importance of getting information faster than others in the market has been a staple in the financial community, since… well… forever. Obtaining a piece of knowledge a little ahead of others allows an investor to get in to or out of market positions before the relevant knowledge impacts general pricing. Entire communication systems have been and deployed, and untold amounts of money have been invested for this purpose. Each wave of innovation in communication systems leads to the inevitable rush to slice some time off of financial relays.

Big Data and its related technologies allow for increases in the speed of communication and the dissemination of information. It, therefore, is the current embodiment of the long-term financial market trend of leveraging advancements in communications to gain information just ever so marginally faster. The best example at present is how social media is being used to displace, or at least buttress, traditional financial news. This is because, in many instances, the millions of everyday people using Twitter will be able to provide information about an event faster than a few thousand professional financial journalists. Traders have discovered that they can receive information nuggets many times faster through something like Twitter than through news channels. The underlying theme in speed-based approaches to Big Data is, “Get me the same information I already like using, just do it faster. I know how to use it and interpret it, as does the rest of the market, so any time gained will be a huge advantage.”

Gaining improved insight though Big Data is just as valuable, albeit less straightforward, than gaining a speed advantage. One of the issues associated with this ‘insight’ approach is that, historically, new forms and types of information must create whole new forms of analysis through which insight can then be gained. For instance, advanced technical analysis of stock trends only became possible when stock price historical records had been kept for a considerable amount of time. Similarly, discounted cashflow analysis only became possible with the advent of reliable accounting information. Finally, macro analysis was only devised through the standardization and availability of macroeconomic data. Arguably, each of these thoroughly different analysis techniques is a cornerstone for much of the investment analysis that currently takes place. It is difficult to envision investing without them. With all of this in mind, it is not unrealistic to imagine that the insights that Big Data analysis provides will become a new type of financial cornerstone in the near future.

A more recent phenomenon, high frequency trading, is also a product of changes in the availability and speed of information. New shorter term tick data, new higher powered computers, and faster, more reliable delivery of automated trades made high frequency trading a reality.

What is next for finance now that data is ubiquitous? If the government’s aggregation of national unemployment, inflation, and other forms of data can revolutionize economic analysis, and if standardizing and regulating quarterly accounting standards and information transparency can revolutionize fundamental analysis, how much more powerful will the transformation of financial analysis be due to the advent of total data?

More Entries:

1 2 3 4

Contact Zettacap

Please leave this field empty.

If you have questions about Zettacap, please contact us. Visit our website if you'd like to enroll in our free beta program.

Legal Notice: