Shift a date by x business days

[lmdamato]

I have a Dataset with two coordinates, event_date and day_offset. event_date is a datetime64[ns], while day_offset is an integer. How can I get the date shifted by the number of business days in the offset?

I tried:

xds["event_date"] + xds["day_offset"]      # offset gets cast to nanoseconds
xds["event_date"] + xds["day_offset"].astype(pd.offsets.BusinessDay)      # numpy.core._exceptions._UFuncBinaryResolutionError: ufunc 'add' cannot use operands with types dtype('<M8[ns]') and dtype('O')
xds["event_date"] + (xds["day_offset"] * pd.offsets.BusinessDay())     # same as above
xds["event_date"] + pd.Series(xds["day_offset"] * pd.offsets.BusinessDay())    # sizes mismatch
xds["event_date"] + pd.to_timedelta(xds["day_offset"]).astype(pd.offsets.BusinessDay)    # TypeError: dtype '<class 'BusinessDay'>' not understood

Is there a way to do this?