High correlation

How would you pick the stocks that are less highly correlated? High correlation refers to when a stock becomes more correlated with the index in a falling market. Example: Stock ABC has a correlation of 0.4, and stock DCE has also 0.4 correlation coefficient with index X. The market fall more than 10% or so in a given period of time and here the correlation of ABC is 0.6, and of DCE 0.86 Indeed ABC is more desirable than DCE as it protects better the portfolio. Anyone knows about a model that allows us to pick the right stock in such situations?

12 Comments
 
SirTradesaLot=correl(return stream 1, return stream 2)

Right, but remember OP that historical correlation does not necessarily imply a future relationship. Traders have been notoriously burned by the market not behaving as expected.

"There are three ways to make a living in this business: be first, be smarter, or cheat."
 
Best Response
Sandhurst
SirTradesaLot=correl(return stream 1, return stream 2)

Right, but remember OP that historical correlation does not necessarily imply a future relationship. Traders have been notoriously burned by the market not behaving as expected.

Agreed. To further the point, if you are doing one stock vs. another, it's almost entirely useless. If you have a large group of stocks and are minimizing correlation amongst those stocks, it can be useful (assuming you are dealing with stocks that are actively traded). To be fair, even if you assume that the correlations remain stable (which is a ballsy assumption), you need to make sure you take into account the variance as well if you are trying to minimize overall volatility or do something similar.
 

Fundamentals build statistics not the other way around.

Generally, companies with stable and more predictable cash flows from operations and/or earnings will be that ABC type of equity.

Example of this are companies from utility industry, AKA defensive stocks. You just can not stop using their services/products, reduce perhaps.

I like to use statistics over CFO and Earnings to define which one from a group of defensive stocks is more stable, but you still need to analyze about companies future and threats to this stream of revenue.

The DCE stock would be more a typical commodity/consume related stock.

"What's your number?" "More."
 
rafaelgirottoFundamentals build statistics not the other way around.

Generally, companies with stable and more predictable cash flows from operations and/or earnings will be that ABC type of equity.

Example of this are companies from utility industry, AKA defensive stocks. You just can not stop using their services/products, reduce perhaps.

I like to use statistics over CFO and Earnings to define which one from a group of defensive stocks is more stable, but you still need to analyze about companies future and threats to this stream of revenue.

The DCE stock would be more a typical commodity/consume related stock.

You're talking statistical models used in conjuction with fundamental analysis, which is a sound method, this way you can avoid spurious correlation between stocks, however the very question I posted in the begining is about those statisticals tools. Basically ABC and DCE were picked from a universe of stocks and were easily screened by just plotting their returns against the main index and just using the Correl function provided by any spreadsheet to get the correlation coefficient, simple and straightforward, is there a similar way for finding the stocks with less volatile correlation?

 

Well this is just an idea but if you are only using returns and their statistical measures(probably to create an algorithm?), i would try to:

1 - calculate stocks average monthly or weekly betas , considering them as population samples (30 samples at least. btw thats why is so difficult to use annual betas which would be better, i think).

2 - rank stocks with low correlation to index and low beta stdev from previous step.

"What's your number?" "More."
 

I've just finished reading the doc, a very interesting one, here you can find it: http://www.icmacentre.ac.uk/pdf/financialriskchapter.pdf Using standard deviation or any other tool which is based on the mean or any other weighted average is incorrect as it will cause a distrotion in the estimation of volatility, correlation will suffer even more from such a simplistic method. They do suggest exponential weighted moving average or the GARCH as a superior method. I am trying to build that algorithm, I used to plot several correlations based on moving averages of returns and then dropping one observation and going to the next, I used 20 observations for every one correlation, finally I calculated the standard deviation of the correlations. I remember vaguely the 30 observations issue, also the central theoreme etc... It is not always possible to get as much data as needed unless you move farther into the past and you will be dealing with a completly different period where underlying economic forces had changed dramatically. I am researching the Spanish main index so I can only use the last 3 years data otherwise there will be so much noise.

 

Cupiditate aut ipsam quas consequatur sit eos voluptatem. Velit sunt neque soluta impedit. Corporis suscipit eos praesentium quisquam ipsa.

Quas rem et ut ea. Suscipit quisquam autem nisi at maxime dolores. Et voluptatem enim autem autem voluptatem dolore. Et repudiandae ratione quia sunt dolorem qui accusantium aliquam. Et necessitatibus iusto enim odio tempora dolor.

Aut vero ullam non eum vitae. Quia recusandae molestiae sed labore nihil inventore. Eum aliquid odio alias et modi rerum sint.

In non sint hic sed est quas quia. Sunt qui libero eligendi voluptatum delectus. Nemo voluptatem eius sed ab minus iure.

"What's your number?" "More."

Career Advancement Opportunities

June 2026 Investment Banking

  • Evercore 01 99.4%
  • Moelis & Company 01 98.9%
  • JPMorgan 01 98.3%
  • Guggenheim Partners 01 97.7%
  • Morgan Stanley 07 97.1%

Overall Employee Satisfaction

June 2026 Investment Banking

  • Moelis & Company No 99.4%
  • Morgan Stanley 02 98.8%
  • Evercore 01 98.3%
  • BMO Capital Markets 12 97.7%
  • Banco Santander 01 97.1%

Professional Growth Opportunities

June 2026 Investment Banking

  • Evercore 01 99.4%
  • Moelis & Company 01 98.9%
  • Morgan Stanley 05 98.3%
  • JPMorgan No 97.7%
  • BMO Capital Markets 12 97.1%

Total Avg Compensation

June 2026 Investment Banking

  • Vice President (14) $434
  • Associates (44) $258
  • 3rd+ Year Analyst (8) $210
  • 2nd Year Analyst (22) $179
  • Intern/Summer Associate (13) $156
  • 1st Year Analyst (78) $151
  • Intern/Summer Analyst (73) $101
notes
16 IB Interviews Notes

“... there’s no excuse to not take advantage of the resources out there available to you. Best value for your $ are the...”

Leaderboard

1
redever's picture
redever
99.2
2
Secyh62's picture
Secyh62
99.0
3
BankonBanking's picture
BankonBanking
99.0
4
kanon's picture
kanon
99.0
5
GameTheory's picture
GameTheory
98.9
6
CompBanker's picture
CompBanker
98.9
7
DrApeman's picture
DrApeman
98.9
8
dosk17's picture
dosk17
98.9
9
Betsy Massar's picture
Betsy Massar
98.9
10
Jamoldo's picture
Jamoldo
98.8
success
From 10 rejections to 1 dream investment banking internship

“... I believe it was the single biggest reason why I ended up with an offer...”