It's a piece of LaTeX which represents the defining property of a martingale.
The simplest example of a martingale in this sense is a double or nothing gambling strategy in certain games of chance(*), but a better way of understanding them is that they are purely random processes, which cannot be predicted based on historical observations: if you try to predict their future, your best guess is to duplicate the present, regardless of what you've seen in the past.
(*)wherein one proves that double or nothing fails to help one win: when the strategy has no statistical trend, then there is no advantage from using it. -- $E(X_t|F_s) = X_s,\quad t > s$
As I was not staying in academia, I did not really care and did not fight this. Thus the dissertation was never published anywhere and was quickly forgotten.
I had done an executive summary in English but can no longer find the file; I'd need to draft it again; it's probably worth it... In the long run, we're all dead. John Maynard Keynes
Sinon, tu peux aussi m'envoyer le fichier par email, l'addresse que j'ai indiquee sur ET lors de l'enregistrement est bidon, mais fonctionne. -- $E(X_t|F_s) = X_s,\quad t > s$
