I think there is a lot of crap around about these things. "Shading the rear wing" is a good example. Most canards have the front plane raised up to negate most of this effect so how can the canned calculators really take this into account properly?
When I was building a canard a number of years ago, I checked out every way to calculate the cg that I could find and most were just dumb guesses, some of which would mislead you into very poor choices.
One of the worst was Martin Simon's otherwise excellent book on model aerodynamics. The cg calculated with his method would give you a crash for sure.
There is a lot of stupid ideas ("static margin*" is one) in these theories. The only sure way to know what your model needs is to make a chuck glider. However throwing it is not enough - you should attach a string to it so you can swing it around your head. Not only does this make you look foolish but it gives you a longer time to observe the characteristics with a particular CG location.

*The fancy term for FUDGE FACTOR