NSE Nifty-50: Price-volume correlations
It’s somewhat obvious that the strength of a bull market can be gauged by comparing the trading volumes on “up days” versus those on “down days”. In a bullish uptrend, one would expect the daily market turnover to steadily increase as well, indicating sustained participation by investors. On the flip side, a fast runup in price on low volume would be a negative divergence, i.e. a bearish signal that investors aren’t convinced of the market’s trend.
NSE provides historical turnover data, including (for each day) - number of companies traded, total number of trades, total number of shares traded and total value traded. I ran some quick analysis on data for the past 6 months, and found reasonably good correlation between the Nifty-50 index and the total shares/value traded (0.31/0.50 respectively). No big surprises here.
![](\"http://www.galatime.com/images/2006/nifty_turnover1.png" "Nifty-50 versus Total shares/value per trade")
However, we can also use this data to compute the average number of shares per trade, and the average value per trade, and then look for trends as follows:
![Nifty-50 versus Average shares/value per trade](\"http://www.galatime.com/images/2006/nifty_turnover2.png" "Nifty-50 versus Average shares/value per trade")
It’s interesting that while the average value per trade has remained about the same, the average number of shares per trade has dropped from 2000 to ~1250. Does this indicate increased participation by retail investors versus institutional investors?
Related Posts:
February 2nd, 2006 at 12:53 am
Few other indicators such as advance-decline ratios too are looking very weak. Infact, lack of participation can also be guaged from the fact that its the large caps that have been surging in the past few days. Midcaps and smallcaps seem to be struggling to stay afloat. Not good signs these, isnt it ? Dont you think that the markets are headed for a correction ?
Ravi.
February 2nd, 2006 at 7:25 am
Ravi,
I’m not so sure that the lack of strength in small/mid caps is necessarily a bearish sign. In fact, its usually towards the end of a bull run that the small caps outperform the large caps, as investors buy up everything in sight, including junk small caps. Also, the large caps account for 80-90% of the total market capitalization, and hence the small/mid caps don’t have a strong marginal effect. Even the A/D ratio gets skewed since there are many more small/mid caps than large caps. What would be interesting is to weight this A/D ratio by value/volume. That would be a better indicator, in my opinion.
February 2nd, 2006 at 6:34 pm
Point taken. However, can you help me understand -
1). Why should a poor advance-decline ratio not indicate a weak market ? Or i think the better question would be - what does a poor AD ratio indicate ?
2). Will it be fair to say that the current rally (since Jan) has largely been driven by index stocks (and not necessarily large cap stocks) ?
And if that is the case, dont you think the markets are tiring out and are in desperate need to undergo a healthy (read: sharp) correction, before resuming the journey to the north pole ?
Ravi.
February 2nd, 2006 at 10:40 pm
Ravi,
1) There’s about 700 stocks traded on the NSE. The top-200 probably make up over 90% of the total market cap. Let’s say these top-200 stocks (by market cap) were up and the bottom-500 were down today. The A/D would be 200/500. But is it a negative sign?
2) This is the key question - are a few stocks driving the entire rally in the indices? Unfortunately, the only way to test this is to analyze historical price/volume data for all stocks. I don’t YET have this capability.
As for the market tiring out, who knows? Maybe there’s a bunch of investors wanting to jump in once the Sensex hits 10,000!
February 3rd, 2006 at 8:22 pm
Analysis using Advance-decline ratio is far from precise. If a stock market is coming out of bear, an increaed advaning stocks may indicate a possible bear market. However, you must watch for distribution days in bull market for possible trouble.A distribtion day is marked by advancing volume , yet no price improvement. The market mostly close down in that case. 3-4 days of such distribtion in 3-4 weeks is enough to cause trouble.
For example 5th Jan was a clear distrubtion day. Market went down after that. Jan 17 marked yet another distribution. Jan 31 market tried to peak on poor volume ( a sign of danger). Feb 1 , yet again a distribution. If the market shows distribution of 2 or more days in 2-3 weeks , it would be better to come out of market.No one really know how long and severe bear market may be !
Various stocks advance and decline based on market action and their performance.
Investors business daily(IBD) is a very good source for stock market study.
Regards
Krishna