Table 1 summarizes the results of the various tests. The weakform
informational efficiency of Paris gold markets cannot be
accepted, either for the clandestine (1941–1948) or the official
markets (1949–1955). The only two tests providing mixed results
are the runs test for the US double eagles and the order-1 autocorrelation
test for the napoleon between 1949 and 1955.We attribute
this to the operations conducted by the Banque de France through
the FSC. Indeed, the FSC mostly intervened on the napoleon market.
This may explain why the market in napoleon coins behaves
differently from the others.