Argh! There is an embarrassing …
Argh! There is an embarrassing mistake in the note from 15th/16th, in the consumption-bit on the portfolios (me confusing the discrete-time formula with the continuous-time formula).
The correct consumption-financing portfolio dynamics should be X(t+1)=(X(t)-c(t))(1+r+uV) -- that is, only what you don't consume, should be invested.
The error affects the problem set (#2 and #3), but a quick panicking back-of-the-envelope calculation indicates I should not need to put further assumptions, though a hint for problem 3 should be appropriate.
A new version to be put up soon, but not tonight. My apologies :-(