I've got a question about theory which is probably a one line answer. I use to understand it but I'm stuck right now.
In the Binomial model, we define the progression of the price as:
$$ S_k = S_{k-1} e^{\alpha X_k} $$
where $P(X_k = 1) = p$ and $P(X_k = -1) = 1-p = q$
Now, to determine $\alpha$ and $p$ using the delta hedging argument (rather than risk-neutrality) we define a small time increment $h$ and at time $t_{n-1} = (n-1)h$ a portfolio $\Pi$
$$ \Pi_{n-1} = f(S_{n-1}, t_{n-1}) - \Delta_{n-1} S_{n-1} $$
where $S_t$ is the asset price and $f(S_t, t)$ is the payout. To determine $\alpha$ and $p$ we choose $\Delta_{n-1}$ so that the evolution of $\Pi_{n-1}$ is deterministic. My lecture notes define $\Pi_n$ at $t=nh$ like so
$$ \Pi_n = f(S_{n}, t_{n}) - \Delta_{n-1} S_{n} $$
Substitute $S_k = S_{k-1} e^{\alpha X_k}$
$$ \Pi_n = f(S_{n-1} e^{\alpha X_n}, t_{n}) - \Delta_{n-1} S_{n-1} e^{\alpha X_n} $$
Then we calculate $\Delta_{n-1}$ so that $\Pi_n$ is deterministic etc.
What I do not understand is why do we have $\Delta_{n-1}$ in the formula for $\Pi_n$? Shouldn't the formula for $\Pi_n$ be:
$$ \Pi_n = f(S_{n}, t_{n}) - \Delta_{n} S_{n} $$