LAUSR

Abstract In this paper we propose to study the auto and the cross-correlation between open-close stock market indexes. Our choice took into account the main indexes around the world: North America (Dow Jones and NASDAQ — USA; IPC — MEXICO), South America (BOVESPA — Brazil; MERVAL — Argentina), Asia (Nikkei_225 — Japan; SSE — China; Hang Seng — Hong Kong), Europe (IBEX_35 — Spain; CAC_40 — French; DAX — German; FTSE_100 — England). Thus, for the opening and closing stock market index and its respective return, we applied the DFA method for measuring auto-correlation, as well as, the cross-correlation between these signals with DCCA cross-correlation coefficient. Our results show that for auto-correlation there is long-range (power-law) auto-correlations with: α D F A > 1 . 0 for the original index and α D F A ≃ 0 . 5 for the return. For cross-correlations we found perfect DCCA cross-correlation between open and close indexes at long time-scale, but in short time-scale there are differences between the stock markets. From the point of view of the return, DCCA cross-correlation coefficient between open and close values showing lower DCCA cross-correlation levels. In this case the open returns are not related with the close returns, evidencing the time independence of these observations in this time-scale, for some stock markets.