time series regression data inputs

time series regression data inputs

I try to get a result in regression on LIBSVM I get the same problem on
single feature data or multivariate data
Suppose I try to find the price of sth, and I have the data for N days. I
give the day 1..N-1 price for X as independent variables, and i train it
with the day 2..N price as Dependent variables. So I try to get the next
day's price always.
When I get the solution (Y) on test data, surely I expect the data price
for the days (2..N for the test data). But i realized that the Y values
are always seems like the X values. When comparing the solution Y with X
values (1..N-1) days, results are more succesfull then (2..N) days.
In other words i expect the next day's values but the result is more
succesful on the same day again.
What's wrong with the approach?