2

Stocks are VERY correlated. This is obvious when looking at an index: most stocks chart match peaks and valleys.

What I would like to know is the mechanism behind those correlations?

Is it "buy programs"? Like big money entering simultaneously into many different stocks at once, thus creating the same impulse on multiple stocks? If not, what is it? How does it happen?

EDIT: This questions concerns both stocks in relation to other stocks, and indices. Let's take the s&p500 index and the most heavy stocks in it. They closely move together. WHO is responsible of this, and WHAT happens really at the transactional level, from seconds to seconds for this correlation to happen so strongly?

2
  • Are you asking about the correlation between a stock and an index, or between different individual stocks?
    – BrenBarn
    Mar 7 '20 at 7:31
  • Actually both, please see my edit above Mar 7 '20 at 10:43
1

This is a pretty broad question, but the short answer is simply that the underlying realities of different equities may move similarly on a general basis.

The real value of a stock is that it provides you with a portion of all future cashflows from that company [dividends + perhaps a liquidating payment if the company is ever closed down]. The ability to pay those dividends is based on the company's net income, which you could simplistically say is based on the economy in general, the relevant industry more specifically, and also the direct specifics of that company.

For example, if the US economy worsens, then some people may lose jobs, and total consumer spending decreases - this could impact the value of all consumer product companies. Further, analysts might predict that fewer people will buy cars in the next 2 years; this means that all carmakers might take a particularly large impact. Finally, perhaps a particular automaker, let's say Ford, has just made a massive investment in a new plant, and perhaps that plant is feared to need to close down due to car shortages, so Ford takes the biggest hit of all. At the same time, maybe Tesla continues to attract new buyers, so Tesla might even improve in value over the same period of time.

The point is that some general indicators impact almost every company [if the interest rate decreases, more money is available to, in theory, invest, so more money is invested and stock prices might rise], some indicators impact a lot of companies [like the cruise industry under-performing due to media coverage of trapped tourists], and some indicators are company specific [like multiple executives at a fast food chain all being fired in the same month indicating performance issues].

One final word of caution: you plainly state "most stocks chart match peaks and valleys." Be careful that you are making broad assumptions about how things work and acting on that information. I think you will find that company performance is not as aligned as you believe.

2
  • Yes, there are cases where stocks have their own world and variations. But in case of stocks highly correlated, like major stocks in a weighted indices, what happens at the transactional level for the correlation to happen? Please see my edit in the main question. Mar 7 '20 at 10:51
  • @RobertBrax At the transaction level, the listed price of a stock is simply the last purchase of that stock which occurred. ie: that's the latest price where a buyer and a seller agreed on a price for a stock. When large factors occur impacting all industries / all companies in an index, then trading of all of those items will happen at a new valuation. It may not be 'accurate', but it is 'the market's best interpretation of accurate valuation at that time. Mar 9 '20 at 17:10
1

When you compare stocks to an index that includes them, most are going to be highly correlated. As an extreme example, the charts of most of the 30 DJIA stocks are going to closely resemble the graph of DIA (the Dow ETF). But if you compare those stocks to each other, not necessarily so.

There's an old adage in the market: "A rising tide lifts all boats". In a bull market, you'd expect the share price of all stocks to increase unless there was adverse company news (and vice versa for bear markets). This is particularly exacerbated in volatile market periods like we're seeing now when other than story stocks, everything gets sold off.

But don't assume that everything correlates. Some sectors are often non-correlated (such as precious metals).


In response to your edit, the WHO is easy. Net buy volume moves share price up and net sell volume moves share down. Anyone active in the market contributes to this (investors, traders, hedge funds, institutions, pension funds, etc.).

If you want an example in isolation, read a bit about how Authorized Participants are active in the creation and destruction of ETF shares. If there's a shortage of ETF shares in the market, the authorized participant creates more so in a price weighted ETF like DIA (30 Dow stocks), all of the index constituents will be bought in the same weight. The price movement effect on each stock will be different because the amount of shares on the order book at current price will vary but the general expectation is that over time, this process will raise the price of all of the components unless there adverse news leading to the selling off of a component. There are other situations like program trading but the buy (or sell) distribution is somewhat the same.

1
  • 1
    Great point - comparing an index that aggregates 30 stocks is unsurprisingly going to look a lot like the individual movement in most of those stocks! Mar 6 '20 at 18:59
0

Read this article, it will help you understand correlation and BETA.

Higher beta's usually carry more risk/volatility, but stocks with zero beta doesn't necessarily mean zero risk.

I would also suggest for you to try to solve for beta to understand how risk is calculated, use excel, and pick 4 stocks. Read more below.

https://seekingalpha.com/article/307708-visualizing-beta-stock-risk-and-market-correlation

https://en.wikipedia.org/wiki/Beta_(finance)

https://www.wikihow.com/Calculate-Stock-Correlation-Coefficient

2
  • Thanks for providing method to estimate the correlation level. But I edited my question as what I want to know is not how to compare, but what happens at the transaction level. Mar 7 '20 at 10:44
  • At the transaction level, it's simple as the market moves up, the stock will move in that direction- if it has a positive beta (+1) it will move towards a higher alpha return, and if it's beta is closer to 0, the stock's alpha won't move with the market. Transaction wise, certain stock move up with the market and stay stagnant or opposite to the market. Knowing how correlation works helps you understand what happens at the transaction level. Mar 8 '20 at 20:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.