# zoo time series called ts
# create weekly trend variables
wkgain = diff(ts, lag=5)
#error due to non-numeric argument to binary operator

ts<- zoo(data1,as.Date(dates,"%Y-%m-%d"))
Warning message:
In zoo(data1, as.Date(dates, "%Y-%m-%d")) :
  some methods for “zoo” objects do not work if the index entries in ‘order.by’ are not unique

ts
     Date     Open    High    Low     Close   Volume     Adj.Close
<NA> 31/12/13 3166.26 3169.20 3158.15 3167.43  245359800 3167.43
<NA> 30/12/13 3159.68 3164.86 3150.26 3153.29  107880500 3153.29
<NA> 27/12/13 3144.87 3154.82 3142.37 3149.76  174566000 3149.76

data1
         Date    Open    High     Low   Close     Volume Adj.Close
1    31/12/13 3166.26 3169.20 3158.15 3167.43  245359800   3167.43
2    30/12/13 3159.68 3164.86 3150.26 3153.29  107880500   3153.29
3    27/12/13 3144.87 3154.82 3142.37 3149.76  174566000   3149.76
4    26/12/13 3131.94 3138.15 3130.23 3134.36   61807500   3134.36