Update Previous Price for datatable

Can we assume a time value is present for the busdate? Otherwise, how would you distinguish
1 ,            'ABX', 10/18/2013 ,  7.00  ,   0
1 ,            'ABX', 10/18/2013 ,  9.00  ,   7.00
from
1 ,            'ABX', 10/18/2013 ,  9.00  ,   0
1 ,            'ABX', 10/18/2013 ,  7.00  ,   9.00
?

Sorry, my example should have had the following, since ModelID, Code and BusDate are the primary key

ModelID, Code, BusDate ,     Price,   PrevPrice
1 ,            'ABX', 10/17/2013 ,  7.00  ,   0
1 ,            'ABX', 10/18/2013 ,  9.00  ,   0
1 ,            'ABX', 10/21/2013 ,  10.00 ,   0
1 ,            'ABX', 10/22/2013 ,  8.00  ,   0

In the above example I would the data after the update statement to be
1 ,            'ABX', 10/17/2013 ,  7.00  ,   0
1 ,            'ABX', 10/18/2013 ,  9.00  ,   7.00
1 ,            'ABX', 10/21/2013 ,  10.00 ,   9.00
1 ,            'ABX', 10/22/2013 ,  8.00  ,   10.00

SQL> select * from yourtable;
   MODELID COD BUSDATE        PRICE  PREVPRICE
---------- --- --------- ---------- ----------
         1 ABX 17-OCT-13          7          0
         1 ABX 18-OCT-13          9          0
         1 ABX 21-OCT-13         10          0
         1 ABX 22-OCT-13          8          0

SQL> merge into yourtable y
  2  using
  3  (select modelid, code, busdate, price,
  4   lag(price) over (partition by modelid, code order by busdate) prevprice
  5   from yourtable) x
  6  on (y.modelid = x.modelid and y.code = x.code and y.busdate = x.busdate)
  7  when matched then
  8  update set y.prevprice = nvl(x.prevprice,0);

4 rows merged.
SQL> select * from yourtable;
   MODELID COD BUSDATE        PRICE  PREVPRICE
---------- --- --------- ---------- ----------
         1 ABX 17-OCT-13          7          0
         1 ABX 18-OCT-13          9          7
         1 ABX 21-OCT-13         10          9
         1 ABX 22-OCT-13          8         10

I cannot use lag as it is a 2012 function. I am using sql 2008