Equity risk pricing

Hi,

I have some precisions to ask regarding the following question:

**********************************************************************************************************************************

Equity pricing models assume which risk is priced?

A) Unsystematic. B) Systematic. C) Both systematic and unsystematic.

Your answer: C was incorrect. The correct answer was B) Systematic.

Unsystematic risk can be diversified away. Thus, equity-pricing models are designed to reflect only systematic risk. It is assumed that the portfolio manager will take steps to diversify and reduce risk.

***********************************************************************************************************************************

This might be true for the Beta on the CAPM, but if we use a Fundamental model?

And even the beta, considering how it is calculated, I don’t understand where it takes out the stock’s idiosynchratic risk?

Thank you for those essentials,

Any up on this one?

Have a good one,

You get the beta by regression. The R sq gives you the portion of the risk of the firm that can be attributed to market risks. The balance 1- Rsq is the firm specific risk.

What about how it is calculated? It uses the covariance of two variables-- what they “share”, in a sense. In this case, it’s how the stock returns and market returns move together. Therefore, any movement in stock returns that isn’t shared with the movement in market returns is not factored into the beta calculation (left for the error term).

Well,

You will have the difference of the asset from its mean multiplied by the difference of the market from its mean, this added on a point by point basis and then averaged on the number of points.

Then we root square the result to get the geometric mean of the sum of those multiplyed diffs.

To finish off we baseline by the market variance, this gives us an amplification or reduction of the market movments; one standard deviation of the market normalises the relative movments of the markets and the second one gives us the relative mouvments of the asset’s portion of deviation (yeah, prrrffffffff).

THERE, could be an idiosynchratic portion of variance can be taken as beeing a result of the market move… but not.

Why don’t we apply the beta stright to the market return and not to the premium?

Cause would be correct to covariate the premiums in this case…

And going forward on this we should first asses the idio risk and then extrac the beta…

Anyway, don’t think its relevant for the exam, matter of assumptions

Thx!

Well, Thanks for your answers :slight_smile:

You will have the difference of the asset from its mean multiplied by the difference of the market from its mean, this added on a point by point basis and then averaged on the number of points.

Then we root square the result to get the geometric mean of the sum of those multiplyed diffs.

To finish off we baseline by the market variance, this gives us an amplification or reduction of the market movments; one standard deviation of the market normalises the relative movments of the markets and the second one gives us the relative mouvments of the asset’s portion of deviation (yeah, prrrffffffff).

Thanks for the calculation… I’m asking why you think this doesn’t remove idiosyncratic risk…

THERE, could be an idiosynchratic portion of variance can be taken as beeing a result of the market move… but not. –Then it’s not firm-specific risk if it’s dictated by the market… There is your assumption, its the dictate of the theory: you move together => you do it because the market owns you, systemic. You don’t => There you are idio.

But, yes, my bad, I am questioning the maxime of portfolio th. which we know are not perfect.

Why don’t we apply the beta stright to the market return and not to the premium? Because the market return includes the risk free rate… we want to magnify only the portion of return related to risks associated with equity. If you applied it “straight to the market return” you would also be amplifying the risk-free rate, which makes absolutely no sense (Rf doesn’t vary with risk)… Yes, I got this, but I am using a beta I calculated with the risk free rate, isn’t it? And then… I am applying it free of the risk free rate… problem no?

Cause would be correct to covariate the premiums in this case….The security characteristic line does this… AAAAAA that is better, thx :slight_smile:

And going forward on this we should first asses the idio risk and then extrac the beta… And how do you propose that this is done? You can’t really assess the unexplained variance without knowing the explained variance for a variable…There’s a very simple logic behind regression analysis (a beta estimation technique)– we take observable data for x and y to model a relationship. What is shared between them is encompassed by the deterministic part of the equation (estimated intercept and slopes), and what isn’t shared is handled by the probabilistic part of the equation (the error term– this would capture idiosyncratic risk for a firm in the CAPM). Its unexplained “by the market”, I would go with fundamentals, info, industry and so on…

Anyway, don’t think its relevant for the exam, matter of assumptions – doesn’t seem like a matter of assumpti** ons– it follows logic used in the statistical estimation procedures for beta and other (not highly technical) logic about what beta represents. Assumptions indeed, makes it easy, reliable on a big scale, but highly questionable, anyway thanks a lot for your time.**

Thx!