eventhough this thread is dead I would like to give my solution aswell. I've dealt with the exact same problem. My teacher told me this is because one of my variables is much bigger than the others (In my case: the trading volume was much bigger then the returns in two different moments). Therefore I've divided the biggest variable through 1000000.

data$Volume<-data$Volume/1000000

I hope I have helped someone dealing with the same problem.

I've dealt with the exact same problem. My teacher told me that this is because one of my variables is much bigger than the others. (In my case, the trading volume was much bigger than the returns in two different moments.) The problem is because of limitations in floating-point computations and precision, not anything inherently mathematical or statistical. 

Therefore I've divided the biggest variable by 1000000.

data$Volume <- data$Volume / 1000000