8
$\begingroup$

A stock has beta of 2.0 and stock specific daily volatility of 0.02. Suppose that yesterday's closing price was 100 and today the market goes up by 1%. What's the probability of today's closing price being at least 103?

$\endgroup$
5
  • $\begingroup$ Hi Ginger, welcome to quant.SE! I've removed your 'disclaimer' and cleared up the title. However, I believe one thing is missing: what model are you using? $\endgroup$
    – Bob Jansen
    Aug 20, 2014 at 9:28
  • $\begingroup$ Hi, Rob, thanks! What model should be using here, this is the question I am thinking of. This is an interview question. Since I am new so I thought there is a classical model for this problem, is it? $\endgroup$
    – Ginger
    Aug 20, 2014 at 10:37
  • $\begingroup$ If I can choose the model, I would do it like this, R_t-R_y is normal distribution, R_t is today's closing price, R_y=100*(1+1%)=101. and $R_t-R_y\sim N(0,0.02)$. Then beta of 2.0 would be useless... $\endgroup$
    – Ginger
    Aug 20, 2014 at 10:42
  • $\begingroup$ or should Rt−Ry∼N(0,0.02*2)? $\endgroup$
    – Ginger
    Aug 20, 2014 at 10:46
  • 4
    $\begingroup$ I also did that sample question for a company beginning with G many moons ago ;)! $\endgroup$
    – Chinny84
    Aug 20, 2014 at 20:44

1 Answer 1

11
$\begingroup$

Usually stock returns are assumed to be normally distributed: $R\sim N(\mu,\sigma)$

If market goes up 1%, the expected stock return is $\mu = \beta\cdot 0.01 = 0.02$ (since β is the senstivity to market).

Stock price going from 100 to over 103 requires a return $R$ of at least 103/100 – 1 = 0.03.

As we have from the question σ = 0.02, we get:

$$ P(R\geq 0.03) = 1 - P(R\leq 0.03) = 1 - F(0.03) = 1 - \Phi\left( \frac{0.03-\mu}{\sigma} \right) = 1 - \Phi(0.5) = 0.31 $$

where $F$ is the generic normal cumulative distribution function and $\Phi$ is the cumulative distribution function for the standard normal distribution.

$\endgroup$
3
  • $\begingroup$ Aren't stockreturns R usually assumed to be log-normally distributed? $\endgroup$
    – Meneldur
    Nov 27, 2015 at 16:23
  • 1
    $\begingroup$ @Meneldur The log-returns are normally distributed and the stockprices are lognormally distributed. It follows from the assumed Geometric Brownian Motion where $S_t=S_0e^{rt+\sigma W_t}$. $\endgroup$
    – emcor
    Nov 27, 2015 at 17:15
  • $\begingroup$ Hello, sorry for waking up the question after years .. why is the stock expected return equal to $\beta\cdot 0.01$? $\beta$ links the stock expected return with the market expected return, and 0.01 is today's market return, not the market expected return. $\endgroup$
    – marco
    Nov 14, 2020 at 20:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.