This is a follow up to a post I had done on the 19th of March on inter-market associations for the swing trader, especially the kind who takes his cues from overnight moves in the SPX or the morning moves in the Hangseng or the Taiwan index. You can find my last post on the subject here.
I want to give a shout-out to Manu who follows the Taiwan index and had requested for the post.
That all the world markets are inter-linked is a foregone conclusion (especially for me) and the same can be verified by plotting their charts on google or yahoo finance where a study stretching back several years can show you similar peaks and troughs.
With that knowledge as a background, one can use one market to predict the moves in the other simply by using the popular "revision to the mean" approach.
Here's a chart of the co-relation:
The traditional co-relation factor between the Nifty and the SPX and Hangseng is 90 % whereas for the Taiwan index it is 73 %.
Now if we look at the 50 day average one can argue that the Nifty has underperformed vis-a-vis the SPX and the Hangseng.
The 10 day average shows the pendulam swinging towards the historical average and the 5 day shows a break-down in the Nifty- Spx relation, but not the Nifty-HSI.The chart shows that the Nifty-Hsi and the Nifty-Twii relation has gone back to it's historical average
We can analyse 2 things from the 5 day average data the chart shows :
1) The Hangseng and the Taiwanese indices are highly co-related at the moment to the Nifty ( 5 day average = historical co relation) .One can take cues from these markets for the Nifty session for the next few days.A higer/ lower close there should mean the same for the Nifty.
2) The breakdown between the SPX and the Nifty brings the principle of "revision to the mean" in play.So either the SPX will correct over the next 10 days or the Nifty would rise over the next ten days to ensure balance.
Remember these are only cues to take note of. The best trades will always come from the chart which is open in front of you!
Showing posts with label Co-relation. Show all posts
Showing posts with label Co-relation. Show all posts
Friday, April 23, 2010
Friday, March 19, 2010
Where's the inter-market study dude?
I put up that title to chide myself in doing this post, 'cos this is the reason I started this blog-- to "study inter-market relationships".
I have often encountered hostile reactions in blogs and public forums, when I talk about the US markets or the FTSE or the Hangseng where the discussion is around the Nifty. They say Dow, Cow and Mow do not matter in the Indian context. Really? I wonder whether some of these fellows were born post January 2008!
To illustrate my need for this post, let me put up this chart:

This is a chart of 4 major indices : Nifty, SPX, FTSE and the Hangseng.
It is a foregone conclusion simply by looking at the chart that each of these markets is connected to the other albeit in varying degrees.
And as you can see they are making their moves together....but that is the broader picture.
I'm here to talk about the minor deviations, get a bit more precise and develop edges in different time frames and to use one market to predict moves in the other.
Sometimes one market knows something the other doesn't and that's where the edge develops.
If two markets are in focus, and they move at the same time together, then it does not help, but it is a confirmation move- one for the other. But when these co-relations breakdown one becomes predictive for the other.
This is what I found in the Nifty a while back...
Here is a snapshot of an excel sheet I maintain :

To the right I have closing prices of the indices for the past one year. The box on the left is the co-relation factor with the Nifty.
As you can see, the corelation with the US spx is 91 % and that with the FTSE and the HSI is 90 %. Incidentally Nikkei is the least at 84 %.
Now let's get to the interesting part :
I have plotted the 50 day and the 10 day averages for the co-relation.
If you see the 50 day average fore the Nifty, it is negatively co-related ( had broken down) and in the past ten days there has been a revision to the yearly average.One look at the 50 day average , 10 days back, would have told me that there has to be a big upmove coming.
The co-relation can work the other way around too...the Spx or the other indices can fall to make up with the Nifty, but the odds are better for one index to move up than for four to fall.
In view of the expected drop on Monday in the Nifty on account of the RBI's announcement, one would be advised to keep an eye out on global indices and this co-relation index.
I have often encountered hostile reactions in blogs and public forums, when I talk about the US markets or the FTSE or the Hangseng where the discussion is around the Nifty. They say Dow, Cow and Mow do not matter in the Indian context. Really? I wonder whether some of these fellows were born post January 2008!
To illustrate my need for this post, let me put up this chart:

This is a chart of 4 major indices : Nifty, SPX, FTSE and the Hangseng.
It is a foregone conclusion simply by looking at the chart that each of these markets is connected to the other albeit in varying degrees.
And as you can see they are making their moves together....but that is the broader picture.
I'm here to talk about the minor deviations, get a bit more precise and develop edges in different time frames and to use one market to predict moves in the other.
Sometimes one market knows something the other doesn't and that's where the edge develops.
If two markets are in focus, and they move at the same time together, then it does not help, but it is a confirmation move- one for the other. But when these co-relations breakdown one becomes predictive for the other.
This is what I found in the Nifty a while back...
Here is a snapshot of an excel sheet I maintain :

To the right I have closing prices of the indices for the past one year. The box on the left is the co-relation factor with the Nifty.
As you can see, the corelation with the US spx is 91 % and that with the FTSE and the HSI is 90 %. Incidentally Nikkei is the least at 84 %.
Now let's get to the interesting part :
I have plotted the 50 day and the 10 day averages for the co-relation.
If you see the 50 day average fore the Nifty, it is negatively co-related ( had broken down) and in the past ten days there has been a revision to the yearly average.One look at the 50 day average , 10 days back, would have told me that there has to be a big upmove coming.
The co-relation can work the other way around too...the Spx or the other indices can fall to make up with the Nifty, but the odds are better for one index to move up than for four to fall.
In view of the expected drop on Monday in the Nifty on account of the RBI's announcement, one would be advised to keep an eye out on global indices and this co-relation index.
Subscribe to:
Posts (Atom)
