Cramer’s Stock Pick Recommendations Analyzed

October 16, 2010
By

(This article was first published on R-Chart, and kindly contributed to R-bloggers)


Bill Alpert of Barron’s has demonstrated the use of R in financial journalism as he criticized the performance of Jim Cramer’s stock picks.  Patrick Burns was an advisor for the analysis done in the article.  R was an important tool that allowed them to do their research as indicated by Burns in his article and by Alpert in his presentation and in an article published in R News.  Their specific critique were centered around obtaining objective verification of claims by CNBC that by following Cramer's advice, one could beat the S & P 500 index.   I recommend you to these sources if you are interested in a more comprehensive analysis.

The following is is based upon data available on the Mad Money Stock Screener as of 10/15/2010.

Mad Money Stock Screener
Since Cramer's Mad Money Stock Screener is available on line there is at least an "unofficial" group of Cramer's recommendations available to analyze.  It is apparent that the data is not complete, as several program segments in the drop down are not represented.  Selecting any of the following segments results in no records returned:
  • Caller's Stock
  • Game Plan
  • Sudden Death


In addition, the date provided includes month and day (but not year information).  It appears that data from one calendar year is available through the web site.  This was born out by plotting the price at the time of the recommendation on charts for individual equities.

Always A Bull Market Somewhere
According to his profile, "Jim Cramer believes that there is always a bull market somewhere, and he wants to help you find it".  His optimistic, entertaining and confident approach that he exudes on screen is reflected in his history of stock picks.  Recommendations are indicated either by a number below or by name:

#   Description
5   Buy
4   Positive
3   Hold
2   Negative
1   Sell

Cramer is functioning as an entertainer with financial knowledge.  The show is geared towards providing action oriented advice (buy/sell) and tends to be skewed towards positive actions.  This fits with his profile description - if there is always a bull market, there is always something to buy.  

Analysis of All Recommendations
The following jitter demonstrates that a 5 (Buy Recommendation) is most often given, and the smoothing indicates that Cramer is generally positive in his ratings.

The program segment in use might also shed some light on the recommendations given.  As mentioned earlier, not every program segment is represented in the data.

The specific totals represented above:

     Buy     Hold Negative Positive     Sell 
    2336        7      229      422      559 


The vast majority of the time, Cramer gives a buy recommendation.  The second most often provided recommendation is to sell.  A clear "Buy" or "Sell" is certainly more entertaining to hear than a "Meh...  hold."  None of this in and of itself means that Cramer's ratings are bad or inferior to other sources.  It simply serves to illustrate that the program is geared towards entertainment.  At best, one might hope that only the clear winners are topics of conversation on the program.  However, further analysis at least calls this into question.

Individual Stock Recommendations
Not every company has an equal number of recommendations.  The top 5 (in terms of total recommendations made) are Apple, Citigroup, Intel, Bank of America and Ford Motor Company.  In order to get a sense of the a given recommendation in the scope of wider market history, the data from the stock screener can be superimposed on a stock chart.

Apple

Apple was given a recommendation on 90 occasions (more than twice times as many than the next most popular companies recommended). The average recommendation was 4.933333 and Cramer recommended Buy 84 times and Positive 6 times.  The clear upward trend in Apple's price justifies an optimistic view in recent history.  None of the remaining stocks in the top 5 recommended fits this pattern though.

Citigroup

Citigroup was given a recommendation on 40 occasions and had an average recommendation of 4.850000, Cramer recommended Buy 36 times and Positive 3 times and Negative 1 time.  The negative rating was on the 12/08/2009 Lighting Round.  

Intel


Intel was tied for second place with Citigroup with 40 recommendations.  It had an average recommendation of 4.900000.  He recommended Buy on 36 occasions and Positive on 4 occasions.


Bank of America 

Bank of America had an average of 4.820513 in the 39 times it was recommended.  It was given a Buy recommendation 34 times, a Positive 4 times and a negative once (During the 11/24/2009 Lighting Round).

Ford Motor 
Ford Motor Company was given an average of 4.777778 in the 36 times it was recommended (a Buy 28 times and a Positive 6 times).

A Negative Example
One additional stock that is of interest is British Petroleum (BP), which had an rough ride this year due to the Deepwater Horizon Oil Spill.  



Cramer issued a Sell recommendation on 10 occasions, a Positive on 1 occasion and a Buy on 6 Occasions (a total of 17 recommendations).

Conclusion
I'll leave it to you to draw your own conclusions about how to interpret recommendations given by Jim Cramer.  For myself, I find him entertaining and well informed about financial news for a wide range of stocks.  He also has the track record as a successful hedge fund manager over the course of a number of years.  However, I am skeptical about the ability of analysts to consistently predict the direction of the market.  

Depending upon the reception of this post, I can provide additional information about the methods used to obtain the data and create the charts above and also show how other stocks recommended on Mad Money have performed.  Let me know if you have any interest in the comments.



To leave a comment for the author, please follow the link and comment on his blog: R-Chart.

R-bloggers.com offers daily e-mail updates about R news and tutorials on topics such as: visualization (ggplot2, Boxplots, maps, animation), programming (RStudio, Sweave, LaTeX, SQL, Eclipse, git, hadoop, Web Scraping) statistics (regression, PCA, time series, trading) and more...



If you got this far, why not subscribe for updates from the site? Choose your flavor: e-mail, twitter, RSS, or facebook...

Comments are closed.

Top 3 Posts from the past 2 days