CAPM_diversification(positively correlated stock)

[kangkan]

Hi

Is there any condition tat governs weterh diversification will lead to risk reduction in case of two stock iswhich are positively correlated?

[vandita24x7]

Two correlated stocks should never be kept in the same portfolio. There is huge theory behind this in quantitative finance.
Btw wht sudden interest in finance??

vandita

On 19 Jun 2014 22:02, "kangkan [via Discussion forum]" <[hidden email]> wrote:

Hi

Is there any condition tat governs weterh diversification will lead to risk reduction in case of two stock iswhich are positively correlated?

---

If you reply to this email, your message will be added to the discussion below:

http://discussion-forum.2150183.n2.nabble.com/CAPM-diversification-positively-correlated-stock-tp7593095.html

To start a new topic under General Discussions, email [hidden email]
To unsubscribe from Discussion forum, click here.
NAML

[kangkan]

actually i was trying to figure out wether diversifying  labour income wud be good stratergy for a household....

[vandita24x7]

What is ur topic fr research proposal ?

vandita

On 19 Jun 2014 22:13, "kangkan [via Discussion forum]" <[hidden email]> wrote:

actually i was trying to figure out wether diversifying  labour income wud be good stratergy for a household....

---

If you reply to this email, your message will be added to the discussion below:

http://discussion-forum.2150183.n2.nabble.com/CAPM-diversification-positively-correlated-stock-tp7593095p7593099.html

To start a new topic under General Discussions, email [hidden email]
To unsubscribe from Discussion forum, click here.
NAML

[Granpa Simpson]

actually two extreme cases in CAPM are correlation=1 and correlation=-1 which deal with the cases of perfect diversification and non diversification...you can refer to Financial Management by I.M Pandey for further reference.

[Granpa Simpson]

And in fact the efficient frontier in CAPM model is derived using these two extreme cases..!!!!

[Dreyfus]

Hey kangkan, Vandita and Subhayu, I don't think(personally) that correlation is a good measure when considering financial time series especially. As what elementary finance says that if u hv a portfolio of two stocks then always choose those who hv negative correlation so that risk is almost neutralize but is this always a good criterion? I mean if I hv two stocks say one of Indian company and another one of UK firm and furthermore this Indian does not related by any means to UK firm but the correlation between the returns of these stocks is -1. Will u still consider is portfolio?? My answer will be no in any case :p

[kangkan]

@vandita..my topic is compensation for people displaced by eviction.

Can some one help with this-

I have the option of working in 3 markets- Market 1- with cosntant wage rate =w

market 2 - wage rate is normal variable with mean 10 and vriance 100

market 3-wage rate is a normal variable with mean 5 andd variance 50

the correlation coefficient between wage in mkt 2 and 3 is 0.2

i have 1 hour to work .

i allocate x fraction to mkt 2 and y fraction to mkt 3 and remaining (1-x-y) to mkt 1.

I need to find the PDF of the wage earnings

I believe it will involve bivariate normal distribution,but i cant figure out how to to procceed.

thanks :)

[Granpa Simpson]

But won't that depend on the nature of the model that u are setting up..that is weder its an additive model or a multiplicative or a model of any other specification..?

[kangkan]

can you elaborate..by the way is a linear combination of two correlated normal variables always normal?

[Granpa Simpson]

Yes kangkan that's true enough that the linear combination of rv's which are normal will be a normally distributed too..!!!

[kangkan]

is there a  fromula for the mean and vriance..remeber that the coefficient of correlation is P

[kangkan]

assume the linear combination to be aX +bY

[Granpa Simpson]

BTW I don't think if they are correlated it will necessarily be normally distributed, because a linear combination of normal rv's is normally distributed only if the rv's are iid's..!!!

[kangkan]

yes...i think there is general confusion about this..wikipedia says something else and some other forum says something else..i will make my life easier and assume they are independent

[Granpa Simpson]

..he he..dat will help u in ur analysis I guess..nywa I don't know about wikipedia but in most of the books it has been specified that the aforesaid condition will hold only if the rv's are iid's..!!!