At last, we included 5 more months and used the same methodology of our first analysis, reexamine our Null hypothesis based on 6 months data instead of only one month. The results were quite similar to our one-month analysis.