1 files changed, 101 insertions, 0 deletions
diff --git a/doc/users/tutorial.rst b/doc/users/tutorial.rst
new file mode 100644
index 0000000..e0336fd
--- /dev/null
+++ b/doc/users/tutorial.rst
@@ -0,0 +1,101 @@
+.. _tutorial:
+
+=========
+ Tutorial
+=========
+
+In this tutorial, we will demonstrate the basic use of nitime in initializing,
+manipulating and analyzing a simple time-series object. For more advanced usage
+see the examples section (:ref:`examples`)
+
+In order to get started, import :mod:`nitime.timeseries`:
+
+.. code-block:: python
+
+   In [1]: import nitime.timeseries as ts
+
+Then, you can initialize a simple time-series object, by providing data and
+some information about the sampling-rate or sampling-interval:
+
+.. code-block:: python
+
+   In [2]: t1 = ts.TimeSeries([[1,2,3],[3,6,8]],sampling_rate=0.5)
+
+If you tab-complete, you will see that the object now has several different
+attributes:
+
+.. code-block:: python
+
+   In [3]: t1.
+   t1.at                  t1.metadata            t1.time
+   t1.data                t1.sampling_interval   t1.time_unit
+   t1.duration            t1.sampling_rate		
+   t1.from_time_and_data  t1.t0      
+
+Note that the sampling_interval is the inverse of the sampling_rate:
+
+.. code-block:: python
+
+   In [4]: t1.sampling_interval
+   Out[4]: 2.0 s
+
+In addition, the sampling rate is now represented with the units in Hz:
+
+.. code-block:: python
+
+   In [5]: t1.sampling_rate
+   Out[5]: 0.5 Hz
+
+Also - once this object is available to you, you have access to the underlying
+representation of time:
+
+.. code-block:: python
+
+   In [6]: t1.time
+   Out[6]: UniformTime([ 0.,  2.,  4.], time_unit='s')
+
+Now import the analysis library:
+
+.. code-block:: python
+
+   In [7]: import nitime.analysis as nta
+
+and initialize an analyzer for correlation analysis:
+
+.. code-block:: python
+
+   In [8]: c = nta.CorrelationAnalyzer(t1)
+
+The simplest use of this analyzer (and also the default output) is to compute
+the correlation coefficient matrix of the data in the different rows of the
+time-series:
+
+.. code-block:: python
+
+   In [9]: c.corrcoef
+   Out[9]: 
+   array([[ 1.        ,  0.99339927],
+          [ 0.99339927,  1.        ]])
+
+but it can also be used in order to generate the cross-correlation function
+between the channels, which is also a time-series object:
+
+.. code-block:: python
+
+   In [63]: x = c.xcorr
+
+   In [64]: x.time
+   Out[64]: UniformTime([-6., -4., -2.,  0.,  2.], time_unit='s')
+
+   In [65]: x.data
+   Out[65]: 
+   array([[[   3.,    8.,   14.,    8.,    3.],
+           [   8.,   22.,   39.,   24.,    9.]],
+
+          [[   8.,   22.,   39.,   24.,    9.],
+           [  24.,   66.,  109.,   66.,   24.]]])
+
+
+
+   
+