Select the *Time
Series *tab of the *SANN
- Data Selection *dialog box to access the options described here.
This tab is only available for time series analysis types.

In time series problems, the objective is to predict (later) values of a variable or variables (steps ahead), from a number of (earlier) values of the same or different variable or variables (steps used to predict). In the most common case, a single variable is involved, and a number of sequential values are used to predict the next value in the same sequence (Bishop, 1995). In that sense, earlier cases in the target variable(s) actually become inputs which are used to predict the variable itself.

*SANN* supports a more
general model: the input and target variable(s) do not have to be the
same, and the prediction can be more than one step ahead.

**Steps.** Use the options in this group
box to specify certain information about the steps used in the time series
network.

**Number of time steps used
as inputs.** In this field,
specify the number of lagged time series values to provide as input to
the network.

**Number of steps ahead
to predict.** In this field, enter the number of time steps ahead of
the lagged input values that the predicted output lies.

For example, to predict a
variable at time t, from lagged values at t-3, t-2 and t-1, enter a *Number
of steps ahead to predict *of
*1* and a *Number of
time steps used as inputs*
of *3*. See handling of data in time series analysis for more
details.

**Note:**
Most applications of time series analysis use *Number of steps ahead
to predict 1*, and the input variables are identical to the target
variables (usually, there is only one variable). In this case, the