Intermarket Study

uasish · #21 3rd June 2007, 11:30 AM

Sunil,
Plz pass on any good link to learn the basics in a simple way.

Asish

beginner_av · #22 3rd June 2007, 12:41 PM

Quote:

Originally Posted by CreditViolet

Spurious Correlation - Correlations that do not imply causation

elementary my dear Sir! good see you on the forum again. A r square may clear up some stuff, but taking all instruments just to find cor may be a futile effort even if we find one! your comments please Sir!

Linus · #23 3rd June 2007, 12:58 PM

Quote:

Originally Posted by uasish

Sunil,
Plz pass on any good link to learn the basics in a simple way.

Asish

Yes I will, but what I have right now is a hard copy.

ss

beginner_av · #24 3rd June 2007, 02:35 PM

Quote:

Originally Posted by CreditViolet

Spurious Correlation - Correlations that do not imply causation

Correct me if I am wrong. I thought that no cor implies causation...it just says how high or low the relationship is.

CreditViolet · #25 3rd June 2007, 09:06 PM

Quote:

Originally Posted by beginner_av

Correct me if I am wrong. I thought that no cor implies causation...it just says how high or low the relationship is.

Lets look at this with an eg.
The sun rises when the rooster crows.There is a heavy correlation between the rooster's crow and the sun rising. So one can assume that the rooster's crow leads to the sun rising and indeed if it does then can one make the sun rise earlier by making the rooster crow earlier?

The subject of correlation is tricky for eg correlations fall apart when the markets are falling and to just calculate one-sided correlations might be tricky.Here cointegration is far more useful than using direct correlations.

PS - I didnt get time to answer on ur FDI PM, I need to reply in a bit detail since I think the whole cycle determination thing is barking up the wrong tree.

Amit Samar · #26 3rd June 2007, 09:17 PM

Hello

every one

kkseal · #27 3rd June 2007, 10:46 PM

Quote:

Originally Posted by uasish

"Spurious Correlations"
Correlations that are due mostly to the influences of one or more "other" variables. For example, there is a correlation between the total amount of losses in a fire and the number of firemen that were putting out the fire; however, what this correlation does not indicate is that if you call fewer firemen then you would lower the losses. There is a third variable (the initial size of the fire) that influences both the amount of losses and the number of firemen. If you "control" for this variable (e.g., consider only fires of a fixed size), then the correlation will either disappear or perhaps even change its sign. The main problem with spurious correlations is that we typically do not know what the "hidden" agent is. However, in cases when we know where to look, we can use partial correlations that control for (i.e., partial out) the influence of specified variables.
So CV has expressed his opinion that my method of trying to find a correlation with indicator MACD & the underlyings close ,according to him "Spurious Correlations" ,a statistical term, & becoz of my lack of knowledge i thought the word Spurious is meant to bash me,(as per HariOm hitting sixers).
Thks CV for channelling my efforts to get deeply involved in the subject of correlation.

Not factoring any 3rd (or other) variable/s can produce spurious results.

Like take Sunil's observation on Metal & Auto Indices. Common sense dictates that a rise in metal (steel particularly although aluminium is also used for engines & other components) prices would result in metal index moving up & auto index moving down i.e. an inverse relation; but there's a 3rd factor - the interest rate scenario.

So there was a time when steel prices were moving up but the Auto index wasn't doing too badly as the interest rates were down. This was followed by a phase when steel prices moved down while interest rates hadn't firmed up yet; this was the best phase for Auto stocks. Next int rates started moving up but steel was still depressed (or just about stabilized) and Autos continued to do well (though the momentum decreased). Now we have a situation when interest rates have mounted and metals also rising - the worst scenario for Autos.

From the above it seems that the correlation between Auto & Int rates is stronger than that between Auto & Metals. In any case you cannot get a clear picture without factoring in all three.

Another variable would be oil prices - but the impact of this would be less in a controlled pricing regime like in India (where oil comps bear most of the brunt), while in say the US this might be the biggest factor influencing Auto sales.

Regards,
Kalyan.

uasish · #28 3rd June 2007, 10:49 PM

Presently this is not my area of priority.

kkseal · #29 3rd June 2007, 11:01 PM

Although cause-effect isn't a necessary implication of any correlation, nonetheless, shouldn't there be some logic in WHAT we are trying to correlate? I mean what's the point in trying to correlate the increase in population to an increase/decrease of the sea levels? Trying to correlate the latter with global warming might be more insightful (although increase in population might have some correlation with global warming - greater fuel consumption etc. - but the main point here is that it has to be meaningful).

Also which one is the cause & which one is the effect may also be debatable in some situations, but it should be able to produce a clear insight on the effect of altering one on the other - otherwise it's a meaningless exercise imo.

Regards,
Kalyan.

kkseal · #30 3rd June 2007, 11:07 PM

When everything rises and none of the correlations work => mkt is peaking
When everything falls and none of the correlations work => mkt is bottoming

Regards,
Kalyan.

Similar Threads for: Intermarket Study
Thread	Thread Starter	Forum	Replies	Last Post
Intermarket Study	Linus	Technical Analysis	2	11th November 2006 01:42 AM
Coal Is Equal To Gold Study	vishalgnu	Energy	0	28th September 2006 03:38 PM
comparative study of Trading Platforms	bhangima	Brokers & Demat Matters	7	12th May 2006 07:01 PM
Way to a good study site	royapa	Technical Analysis	2	5th September 2004 11:01 AM