Nifty has 50 stocks and each of them have different weight age. Just 3 stocks, Reliance, Infosys and ICICI have about 25% of weight in index price computation.
This makes the index tilt more towards the price movements of the these stocks. What about other quality stocks whose price movement has less effect on index. What if we increase there weight so as to better reflect the movement of broad stocks. Well there is such a way of index computation called equal weight where all the constituents have equal weight.
The S&P500 equal weight index returns completely overshadows the S&P500 index.. See the chart below:
The equal weight index has completely beaten the SPX over last 10 years and has outperformed by a factor of 20 times. Astonishing.
What if we do the same study for Nifty. As the equal weight index data for Nifty is not available we can have to make the custom index incorporating all the change in stocks.
The calculation is very tiring as we have incorporate the changes of stocks. Barath the excel guy made this possible. Many thanks to you dude.
Lets plot the Nifty and equal weight index from 2005 onwards.
The index has not outperformed the Nifty. What does it signify. To me it seems that the top 5 constituents have outperformed Nifty while the bottom 10 has underperformed the Index from 2005 onwards.
This makes the index tilt more towards the price movements of the these stocks. What about other quality stocks whose price movement has less effect on index. What if we increase there weight so as to better reflect the movement of broad stocks. Well there is such a way of index computation called equal weight where all the constituents have equal weight.
The S&P500 equal weight index returns completely overshadows the S&P500 index.. See the chart below:
The equal weight index has completely beaten the SPX over last 10 years and has outperformed by a factor of 20 times. Astonishing.
What if we do the same study for Nifty. As the equal weight index data for Nifty is not available we can have to make the custom index incorporating all the change in stocks.
The calculation is very tiring as we have incorporate the changes of stocks. Barath the excel guy made this possible. Many thanks to you dude.
Lets plot the Nifty and equal weight index from 2005 onwards.
The index has not outperformed the Nifty. What does it signify. To me it seems that the top 5 constituents have outperformed Nifty while the bottom 10 has underperformed the Index from 2005 onwards.
Fantastic analysis, Sirji. You blog is truly different and a cut above the rest. The frequent experiments you undertake is surely enriching for the readers.
ReplyDeleteCheers!
Btw, 20 times is really a mind boggling no. for a stock market that has had a "lost decade".
ReplyDeleteHey Vineet,
ReplyDeleteThanks for the encouraging words. It keeps me going.
Maybe this study will work better for indices with large member number of member constituents. NASDAQ with 100 can confirm this hypothesis. At the same time if Dow gives flattish returns on equal weights for the same period then yes may be we have a tradable idea. Further if it true then need to check if it can replicated in other equity markets. India has Nifty with 50, Sensex with 30 and CNX100 and BSE 500.
ReplyDelete