Scraping SeekingAlpha: Are Analyst-recommended Stocks Really Outperforming?

Posted on Feb 17, 2018



As a stock investor, I've always wondered if relying on research conducted by stock analysts and industry experts will help me pick the stocks that will outperform the market. Everyday we see thousands of articles capitalizing on emerging trends in the market and pitching on investment ideas. But after the dust settles, how good is their advise? Going into this project. I had these questions in mind: Are stock analysts/industry experts really better at picking stocks that will outperform the market than we do? Are they better at giving ‘long’ or ‘short’ recommendations? Are institutions reports more reliable than the ones issued by individuals?


Data Collection

To study this, I scraped a popular website called seeking alpha. It’s a platform for stock analysts to post research reports and analysis pieces as well as investment ideas and strategies. Both independent analysts and institution researchers can contribute to the contents of this platform. In contrast to other equity research platforms, insight is provided by contributor base of investors and industry experts (buy side) rather than sell side.

I scrapped both long and short ideas from the platform to get a more complete picture. I scraped the title and link to the research articles, the recommended tickers, time of the report as well as the author of the report. Also as step2, I used a package found on Github to query stock closing prices within a one year time horizon for all the tickers I collected in the previous step.

So at the end of the process, I collected info from 2500 stock research reports issued within a period of three months. These articles made recommendations (either buy or sell) on about 1200 different stocks. I also have the price data of one year for all the stocks.


Long vs. Short Portfolios

For the first step in my analysis, I constructed a long portfolio using all the stocks that were recommended buying by analysts on Seeking Alpha, and a short portfolio using all the stocks recommended selling by the analysts. Here's the performance of the two portfolios for last year under different holding periods:

As we can see from the Box plot above, contrary to what the analysts were saying, the short portfolio, which consists of stocks that analysts recommend selling, actually outperformed the long portfolio under various holding periods! It should be the other way around if most analysts were correct about their ratings.


Top 20 Long vs. Top 20 Short

Some may ask, since a stock will be included in the previous long or short portfolio as long as it was recommended once by some random analyst. What if we only want to include the stocks recommended multiple times by multiple analyst? Will our portfolio perform better using this criteria? Let's find out.

This time I constructed the long portfolio using only the top 20 most recommended buying stocks, and the short portfolio using only the top 20 most recommended selling stock.

As we can see from the chart above, our top-20-short portfolio still performs better than the top-20-long portfolio. And more interestingly, the top-20-long portfolio is actually underperforming all of the major stock market indexes (S&P500, Nasdaq, DowJones) for the past year. So it seems that we are disappointed by analysts' recommendations again.


Institution vs individual:

So what about institution research vs. individual research? Within our dataset, institutional researchers contributed about 1/3 of all articles, while individual analysts contribute to the other 2/3. Most people would probably think that research conducted by a institution is more prestigious and thus carries more weight. But is that thinking justified? Are institutional recommendations more accurate?


And after plotting the percentage of stocks that outperform market index (S&P500) within each group of stocks (Individual Long, Institution Long, Individual Short and Institution Short), we don't see institutional reports having an edge over individual in terms of accuracy in predicting outperformances.



During the time frame I chose to conduct this study on, none of the portfolios I constructed using stocks recommended by the analysts outperformed the market. Simply purchasing an index fund that tracks S&P500 or Nasdaq would have yielded meaningfully higher returns. I am hoping that this study would  show to millions of retail investors that relying solely on expert opinions when it comes to picking stock may not be the best strategy. What's more, blindly trusting reports issued by prestigious institutional researchers may not help investors pick the right stocks to outperform the market. As a prudent investors, we should conduct more due diligence, research, while  combining with expert opinions before making an investment decision.

Thanks for reading my blog post and don't hesitate to reach out if you have any questions or feedback on my approach and techniques.

My Github repository for this project here.

Link to my presentation slides here.



About Author

Yinan Jiang

Yinan recieved his Bachelor's Degree in Accounting from Shanghai University of Finance & Economics and Master's Degree in Economics from USC. Before data science, he worked both as an Equity Analyst and Data Analyst for major financial institutions...
View all posts by Yinan Jiang >

Related Articles

Leave a Comment

No comments found.

View Posts by Categories

Our Recent Popular Posts

View Posts by Tags

#python #trainwithnycdsa 2019 airbnb Alex Baransky alumni Alumni Interview Alumni Reviews Alumni Spotlight alumni story Alumnus API Application artist aws beautiful soup Best Bootcamp Best Data Science 2019 Best Data Science Bootcamp Best Data Science Bootcamp 2020 Best Ranked Big Data Book Launch Book-Signing bootcamp Bootcamp Alumni Bootcamp Prep Bundles California Cancer Research capstone Career Career Day citibike clustering Coding Course Demo Course Report D3.js data Data Analyst data science Data Science Academy Data Science Bootcamp Data science jobs Data Science Reviews Data Scientist Data Scientist Jobs data visualization Deep Learning Demo Day Discount dplyr employer networking feature engineering Finance Financial Data Science Flask gbm Get Hired ggplot2 googleVis Hadoop higgs boson Hiring hiring partner events Hiring Partners Industry Experts Instructor Blog Instructor Interview Job Job Placement Jobs Jon Krohn JP Morgan Chase Kaggle Kickstarter lasso regression Lead Data Scienctist Lead Data Scientist leaflet linear regression Logistic Regression machine learning Maps matplotlib Medical Research Meet the team meetup Networking neural network Neural networks New Courses nlp NYC NYC Data Science nyc data science academy NYC Open Data NYCDSA NYCDSA Alumni Online Online Bootcamp Online Training Open Data painter pandas Part-time Portfolio Development prediction Prework Programming PwC python Python Data Analysis python machine learning python scrapy python web scraping python webscraping Python Workshop R R language R Programming R Shiny r studio R Visualization R Workshop R-bloggers random forest Ranking recommendation recommendation system regression Remote remote data science bootcamp Scrapy scrapy visualization seaborn Selenium sentiment analysis Shiny Shiny Dashboard Spark Special Special Summer Sports statistics streaming Student Interview Student Showcase SVM Switchup Tableau team TensorFlow Testimonial tf-idf Top Data Science Bootcamp twitter visualization web scraping Weekend Course What to expect word cloud word2vec XGBoost yelp