Correlation Study on Nifty Stocks

The correlation between any two stocks (or sets of variables) summarizes a relationship, whether or not there is any real-world connection between the two stocks. The correlation coefficient will always be between -1 and +1. These two extremes are considered perfect correlations. A negative coefficient means that the two stocks,, will move in opposite directions (if one stock increases, the other will decrease); a positive coefficient will mean that the two will move in the same direction (as one increases, the other will increase). And if Correlation is zero(perfectly uncorrelated) or almost zero then both the stock will behave independently

About Correlation Table

Correlation table is a two-dimensional matrix that shows correlation coefficient between pairs of securities. The cells in the table are color-coded to highlight significantly positive and negative relationships.

correlation-table
(Click on the Picture for Big Table)

Here i had taken randomly some 15 stocks from the Nifty 50 Pack with the sample duration between 1st Jan 2009 – 14 Dec 2009 and the correlation table is presented above. Also Highly correlated pairs (Both Positive & Negative ) and Zero Correlated Pairs are identified from the above table
An Investopedia article explains how USD is correlated with the US Indices

If we compare the US Dollar Index (USDX), an index that tracks the value of the U.S. dollar against six other major currencies, and the value of the Dow Jones Industrial Average (DJIA), Nasdaq and S&P 500 over a 20-year period, the correlation coefficient calculated for the USDX versus the DJIA, Nasdaq and S&P 500, is 0.35, 0.39 and 0.38, respectively. Note that all of the coefficients are positive, which means that as the value of the U.S. dollar increases, so do the stock indexes, but only by a certain amount. Notice also that each coefficient is below 0.4, which means that only about 35% to 40% of the stock indexes’ movements are associated with the movement of the U.S. dollar.

List of High Positive Correlated Pairs

List of High Negative Correlated Pairs

Zero Correlation Stocks

Correlation can also be used for Portfolio Diversification. An investor can reduce portfolio risk simply by holding instruments which are not perfectly correlated. In other words, investors can reduce their exposure to individual asset risk by holding a diversified portfolio of assets.

Related Readings and Observations

  • Historical Correlation Amibroker AFL code We recently decided to see how correlated the S&P Nifty is with Dow Jones (US index) and if they were becoming more or less correlated in recent years. The chart shows the 20 day […]
  • Nifty and Bank Nifty 90 min charts for 17 Feb 2012 Trading Nifty and Bank Nifty 90 min charts are in buy mode with supports coming around 5396 and 10504. Shorting is recommended on the support breaks.
  • A Fact About Indian Stock Market Dow has corrected only from 14k to 11.5k maxbecause of its own dometic reasons.thats a 17.8% fall from lifetime highs!But Indian Markets from 21206 to 15336! thats a 27% cut,FOR NO VALID […]
  • Simple Rules to Trade Using 5 EMA(LOW-HIGH) USE EMA -5(LOW) and EMA-5(High) instead of using 5 EMA. Rules If the stock/Index trading above 5 EMA(High) - trend is expected to be up with support at 5 EMA(High)and next level of […]

Comments

  1. ss says

    Usually, it is seen that stocks from the same industry are highly correlated. Maruti and Hero Honda are highly correlated at around 0.95. Another example can be say Bharti, RCOM.ss

  2. cas says

    Hello,
    Kindly share the code of “Correlation Study on Nifty Stocks” and guide us how to use this.
    Thank you

  3. eagle says

    Dear Mr. Rajandran

    That “table” is cooooool.

    I have one idea.
    Do you have any interst in “cointegration”?

    Cointegration seems to be better than correlation in some tradings.
    Could it transfer to “cointegration”?

    Thanks.

  4. Pranali says

    Hi Rajandran ,

    Could you please guide me on how to create that table above? What data do you use to calculate it? Can I use the EXCEL corr() function to do it? What timeframe is most useful ?

    Thanks.

Leave a Reply

Your email address will not be published. Required fields are marked *