|
|
|
|
how to measure correlation in a global market?
|
kerryvan
05-04-2008, 7:30 AM | Post #2514484 |
11 Replies
| -4 |
  |
|
|
As the global market place changes, the investments need to be in markets that don't correlate. Alpha and Beta are terms used to track correlation. When comparing two investments it would be wise to measure how well they correlate. Unfortunately, for funds the alpha and beta are used to track to an index, based for that fund style. Comparing the alpha and beta for an emerging market fund with a US large caps is not meaningful or useful. a web search for 'alpha' and 'correlation' did yield a hit for a site that tracks ETF across the global market. This being said, the true ratio of % in stocks, bonds should include markets that don't correlate. So a portfolio of 100% stocks, in the right mix would yield a higher return for less risk than one with bonds.
|
Related Topics
betacorrelationemerging marketFundsstocks
|
Re: how to measure correlation in a global market?
|
xejir
05-05-2008, 6:37 AM | Post #2514791
| 0 |
  |
|
Check out FARRELLs articles on marketwatch.com? I collect examples of various portfolios based on what are considered the best combinations of risk vs reward globally and sector/market cap wise.. Follow the links.
With some you can find good long term research. Aronson and Armstrong are interesting. Search on modern portfolio theory?
|
Related Topics
portfolio
|
Re: how to measure correlation in a global market?
|
pmolloy
05-05-2008, 7:03 AM | Post #2514793
| 0 |
  |
|
i am intrigued re: investing in Vietnam. any advice? is there a fund or ETF dedicated to vietnam?
|
Related Topics
Funds
kerryvan
05-05-2008, 11:22 AM | Post #2514851
| 0 |
  |
|
from a search engine.. I replied becasue I'm interested as well... k http://markets.ft.com/ft/tearsheets/performance.asp?s=XFVT:LSE http://finance.google.com/finance?q=LON:XVTD http://investing.businessweek.com/research/stocks/snapshot/snapshot.asp?symbol=XVTD.L The way how to invest in Vietnam is for example via Vietnam Opportunity Fund (VTOPF.PK). Enter Cavico Corporation (CVIC.ob), the first Vietnamese investment opportunity to list on the American exchanges. American exchanges.
|
| 1 |
  |
|
I could be wrong but according to my knowledge, Alpha - excess return over expected risk adjusted return. Beta - systematic risk (the one that cant be diversified). if its 1 that means it moves in tandem with the market, over 1, more risky than the market, lower than 1 less risky than the market. However, risk is defined as variability of returns, meaning higher risk stock (by higher beta) doesnt mean higher business risk in this context but rather higher dispersion of returns when compared to the market. Therefore now that we know that, we know that alpha has nothing to do with correlation. Beta does in a way because it tells you if it moves along the market or not. For example, gold has negative beta meaning it moves in opposite direction of the market.
Let me know if its wide off the mark.
|
Related Topics
stocks
|
Re: how to measure correlation in a global market?
|
pining4Lenore
05-06-2008, 3:28 PM | Post #2515239
| -1 |
  |
|
"Non-correlative" stock markets do not exist -- and if there are a few that do exist (i.e. Vietnam, Kazakhastan, Outer Mongolia, or Hyperborea) you would probably not want to put a significant portion of your funds in them. I mean really, what percent of your portfolio are you really going to want to put in "frontier" stock markets? When the U.S. stock market goes down hard and long, all major markets come along for the ride. And vice-vera. In fact, correlations RISE when stock markets are in bear territory. Exhibit A: check out the returns for the SP500 and EAFE during the 00-02 bear. Both indices were down about 47% (from peak to trough). Non-correlated ASSET CLASSES are another matter. Here the major 'suspects' would be commodities, short-term bonds, non-US dollar bonds, and that old fave: Cash.
|
Related Topics
bearcommodities
|
Re: how to measure correlation in a global market?
|
kerryvan
05-06-2008, 4:10 PM | Post #2515247
| 0 |
  |
|
The comparision between the us market and EAFE are comparing two rather robust developed markets. I'm trying to understand the comparision between "frontier" markets and developed nations markets. The term emerging markets is way too broad a term, depending on who you talk to it includes the BRIC nations as well as 'frontier' nations. BRIC nations are behaving more like developed markets in recent yrs, The don't have the currency and political exposure that was the primary concern of the 80's to 2000 for emerging markets. Developed countries were seeking the emerging markets for low cost goods in the 80's to 2000. Now the paradigm is shifting that the BRIC countries are shifting to outsourcing to lower cost countries. This shift as well as internal consumption is why the China market is going to have continued growth of 10% while the rest of the developed nations will only have 3-5% growth. I'm looking for a methodology to compare the value proposition that developed markets (US and Europe) vs BRIC vs "frontier" for consideration of investments.
|
Related Topics
ChinaCurrency
|
Increasing Correlations in Down Market-
|
SpringChange
05-17-2008, 9:29 AM | Post #2518845
| 0 |
  |
|
|
When correlations increase, they must also decrease just as fast during down markets, due to the law of magnitude.
|
Related Topics
markets
|
Re: Increasing Correlations in Down Market-
|
SageNolos
05-17-2008, 12:30 PM | Post #2518904
| 1 |
  |
|
pining4Lenore:When the U.S. stock market goes down hard and long, all major markets come along for the ride. And vice-vera. In fact, correlations RISE when stock markets are in bear territory. SpringChange:When correlations increase, they must also decrease just as fast during down markets, due to the law of magnitude.
Your statements contradict. I don't buy either claim. Re: how to measure correlation in a global market? kerryvan:Alpha and Beta are terms used to track correlation. When comparing two investments it would be wise to measure how well they correlate. Unfortunately, for funds the alpha and beta are used to track to an index, based for that fund style. Comparing the alpha and beta for an emerging market fund with a US large caps is not meaningful or useful.
Kerry, (R-squared) R^2 is used to measure correlation. Not Alpha & Beta. As you said, comparisons with them are "not meaningful or useful". A high R^2 of say: 90 means that about 90% of the variation can be "explained" by movements in the other market/index. The lower the R^2, the less they are correlated. Further, a R^2 of less than about 70%, also means that the associated Alpha & Beta figures are basically unreliable (and therefore should be ignored)!
|
Related Topics
Fundsbear
|
Re: Increasing Correlations in Down Market-
|
kerryvan
05-17-2008, 12:50 PM | Post #2518912
| 1 |
  |
|
Okay I didn't word it very well. my fault, What I wanted is some ideas on how to measure market correlation between different foreign markets. The fund alpha and beta are measured against a market index for that fund. You are implying all markets are measured against the same index, which they are not! Europe and China do not use the same index. So my question is: how to measure correlation on a global basis. I'm not sure I agree with the comment when US goes down all others follow. I graph each of the funds I'm in and the shape and peaks/ dips are not the same. Granted US stocks have very similar shapes, but compared to some foreign markets they differ. There are events that will ripple around the globe a couple days, taking the markets down, but then the rebound will be different. So, I can continue to compare the graphs or look for another means, hence the question.
|
Related Topics
FundsChina
| 1 |
  |
|
|
If the statement that correlations increase during down markets is right, then it must be the case that correlations decrease during an up market. And the rate of divergence will be directly dependent upon how much convergence was originally obtained, the law of magnitude is something like a stock falls fifty percent, it must increase one hundred percent, another drops twenty-five percent, it only needs to increase thirty-three percent, all things being equal with two stocks with different betas. Which stock would you own? I'll take the one that triples the return of the other and sit it out, and, due to behavioralism, I'll be in the minority. One more thing, you don't have to buy nothing here, there is no charge for academic gobbledygook especially mine.
|
Related Topics
stocksmarketreturns
xejir
05-26-2008, 7:51 AM | Post #2521650
| 0 |
  |
|
Find a source for some good historical data...Yahoo?...needs to be corrected data. Put it into a spreadsheet...do your own statistical analysis? Or...check this out? http://www.fundsinteractive.com/21st/index.html ...graphics don't show up?
|
|
|
|