Discussion on finance concepts/problems for people interested in finance

i_m_listening

Quote:

Originally Posted by antiquity

Originally posted by Pirate:
I have one doubt about CAPM model for calculating cost of equity of any company..
CAPM sates Ke=RFR+beta*(Expected market return-RFR)
Can anybody please tell me what exactly is expected market return and how to calculate it ????
Is it arithmetic mean of historical returns of market or any other thing ???
TIA

Yes the market return are based on historical averages. For the purpose of calculations you usually take a broad based indices which is representative of market returns eg: Nifty, CNX Midcap, Dow jones industrial avg.
IN support of the above ans, i would like to add that beta is also based on historical figures and hence the useage of historical market return cannot be expected to vetiate the results.

Cheers
ashish

return on market would be calculated on an historic basis and usually on a long-term horizon - 3 yrs or more but annualized and the duration is same as that used for the calculation of beta (beta = cov(mkt,stock)/std dev of the mkt)

return on market would be simply - ((index value t1/ index value t0)^(1/number of years)-1)

you could use any of the broader indices that as u said should be representative of the entire mkt

pirates_???

Thanks for replying friends....
But only the arithmetic caluclations dont work here and need to consider a lot of other factors which i am still finding .....
Any insights are welcome....

@i_m_listerning
If u calculate nifty returns for last 5 years using method u said ( which is right but have not considered many other factors) will give u answer somewhere around 36% and if u put these values in CAPM the figures come too high.... So i m still not sure about it......

i_m_listening

Quote:

Originally Posted by pirates_???

Thanks for replying friends....
But only the arithmetic caluclations dont work here and need to consider a lot of other factors which i am still finding .....
Any insights are welcome....

@i_m_listerning
If u calculate nifty returns for last 5 years using method u said ( which is right but have not considered many other factors) will give u answer somewhere around 36% and if u put these values in CAPM the figures come too high.... So i m still not sure about it......

pirates.. that's y I said that you need to annualize the returns..

so if NIfty returns are 36% over a 5 yr period

annualized return on Nifty would be = ((1+.36)^(1/5))-1 = .0634 = 6.34%

hope this is helpful

pirates_???

See I have calculated GM of return of each year for last 5 years and that answer is 36%.....
Also Nifty mar 03 984 and Nifty mar 08 4734 => [(4734/984)^(1/5)-1]=36%
Do I need to again annualise them ???? Isnt it a double work ????
TIA

GrimMan

Puys, I hope I don't sound like the devil's advocate here, but there seems to be a small mixup. The questions is "How do I calculate the cost of equity of a company? Is it RFR+..."

Here, there are two aspects to be considered.
1. To the owner of the question - Are you using CAPM to get at a buy/sell decision of a company's shares of stock? If so, the formula suggested is appropriate. However, the values you apply for RFR and expected return are all important and here's where the strength of an analyst comes out. The time period of the data you use to arrive at these values for substituting in the formula depends on your analysis requirements. There's no fixed tenure or benchmark.

By applying this formula, using CAPM, you arrive at the best return expected of holding a stock of the same risk profile. If the real return of holding another stock for that time period is less than that predicted by CAPM, it means that either you have not met your cost of capital for holding that stock (i.e another stock carrying the same risk offered better returns) or the values you substituted for the CAPM formula are incorrect.

To do accurate CAPM based calculations, you need to do a sort of regression analysis. For example, substitute Nifty 6 month return, then a 9 month return, then a 1yr return while holding the other values constant and then check for which substituted value, actual returns were closest to the one predicted by CAPM. And this partial derivative or regression analysis should be done every few months as at one point of time, a 9 month return would lead to accurate results. Later a 24 month return would be more accurate.

2. If your question deals with the calculation of cost of equity for a company, its a whole different ball game, where CAPM won't be of much use.

pirates_???

Thanks grimman...
I am (still

) trying to find out cost of equity for a company. I read in some articles we can find out cost of equity using CAPM and so i was trying to use it.
Anyways, I would like to read more about regression analysis if you could post some literature about it.
TIA

GrimMan

Apologies for such a late post. Caught up in the mock test euphoria!
Regression analysis. Am not an expert on this. My understanding is that most economic and business parameters/measures can be estimated with some amount of reliability by framing a polynomial equation. Example - Lets say we have sales and customer data about a brand of luxury cars we are selling. We want to determine how sensitive sales is to a 10% price increase. First we think of a couple of parameters which likely affects customer's demand for cars - Per capita income, disposable income, upper middle class population and price.
W put them in an equation
f(x) = a+b+c+d
F(x) is customer demand; a is per capita income and the others are as mentioned in the line above. Here, we are allocating equal weightage to all the factors. Then we keep b,c and d constant and see how f(x) varies with a - ie how sales varies with percapita income. Here we use correlation and other statistical methods and note that f(x) varies directly with 0.85a.
We do the same for the other parameters (b,c and d) and hence arrive at something like :
f(x) = 0.85a+0.34b+(c/107)-(d/4002.45)

Hence sales is inversely proportional to price. If current values of a,b,c and d are
say, for simplicity's sake 100, 50, 10,70,000 and Rs 10,00,000, a 10% price increase can reduce sales from 9852 to 9827 (a 0.25% drop is sales).
To improve the accuracy of such frecasting models, we can add more parameters to the equation. But this will increase complexity and consume huge resources to chew data and check out the correlation.
In a nutshell, this is regression analysis. Hope this is of relevance to folks.

pirates_???

Can anybody please tell me any good book or links to research papers about forex management in India and also in other major countries or about forex market or valuation of currencies ?

TIA

kk1978

hi,

nice to find a group focussed on cfa - am looking for some information on the CPA course offerred by ICFAI - would need some specifics like what is the student profile & what are the areas the curriculum focusses on . Am a CA INTER - had cleared it 8 yrs back & CA FINALS as a concept has become very stale for me (though i've never even attempted it ) & hence the Accelarated CPA programme seems to make sense .

I am trying to navigate my way into a career in financial reporting & processing for KPOs .. hence also whether the CPA is relevant to my objectives ?

regards....