$X \sim N[m,v^2] => E[X] = m , Var[X] = v^2$
$Y = e^X \sim LogN[m, v^2] => E[Y] = e^{m+0.5v^2} , Var[Y] = E[Y]^2(e^{v^2}-1)$
$\alpha $ = Continuously compounded annual return on the stock
$\overline r $ = Mean of monthly interest rates
$\overline{\sigma}$ = Sample Standard deviation of monthly rates
$\alpha = \frac{\overline r}{h} + \delta + 0.5 [\overline{\sigma}\sqrt{h}]^2$
$n=(\alpha – \delta + 0.5\sigma^2)(T-t)$ | $m=(\alpha – \delta – 0.5\sigma^2)(T-t)$ [btw]* m – minus [/btw]|$v^2 = \sigma^2(T-t)$
$\hat{d_1} = \frac{ln[\frac{S_t}{K}]+n}{v}$ [btw] *Note $d_1 = \frac{ln[\frac{F^P(S)}{F^P(K)}]+0.5v^2}{v}$ and $n=(\alpha – \delta + 0.5\sigma^2)(T-t)$[/btw]| $\hat{d_2} = \frac{ln[\frac{S_t}{K}]+m}{v}$ [btw]*Note $d_2 = \frac{ln[\frac{F^P(S)}{F^P(K)}] – 0.5v^2}{v}$ and $m=(\alpha – \delta – 0.5\sigma^2)(T-t)$[/btw] | $\hat{d_2} = \hat{d_1} – v$
$ln[\frac{S_T}{S_t}] \sim N[m, v^2]$
$lnS_T \sim N[m+lnS_t, v^2]$
$S_T = S_t e^{m+vZ}$
- $ E[S_T] = S_te^{(\alpha-\delta)(T-t)} $ | $Var[S_T] = E[S_T]^2(e^{v^2}-1)$
- $Median = S_te^m$
- $P[S_T < K] = N(-\hat{d_2}) $ | $P[S_T > K] = N(+\hat{d_2})$
- Conditional and Partial Expectation
- $E[S_T | S_T < K] = \frac{E[S_T]N(-\hat{d_1}) }{N(-\hat{d_2})}$ *Note that < and – go together
- $E[S_T | S_T > K] = \frac{E[S_T]N(+\hat{d_1}) }{N(+\hat{d_2})}$ *Note that > and + go together
- x %lognormal confidence interval of $S_T$ => Area under the curve = 1-p = x %
- Calculate p, Calculate $z^L = N^{-1}(\frac{p}{2})$ and $z^U = -z^L$
- Calculate $S_T^L = S_te^{m+vz^L}$ and $S_T^U = S_te^{m+vz^U}$
- $Cov(S_T,S_t) = E[\frac{S_T}{S_t}]Var[S_t | S_0] = e^{(\alpha-\delta)(T-t)} E[S_T |S_0]^2(e^{v^2}-1)$
Note: If the price of a stock can be observed over a time interval, no matter how short the interval is, then $\sigma$ is revealed immediately by determining the quadratic variation of the logarithm of the stock price.
$\underset{n \to \infty}{Lim}\sum _{j=1}^n\left \{ ln[S(jT/n)/S(j-1)T/n)] \right \}^2 = \sigma ^2T$
Financial MathematicsDerivative Markets