Time Series Processes with Initial Conditions 

Mathematica 10 supports time series processes with given past values, which cover both the stationary time series conditioned on past values as well as nonstationary time series with initial conditions. When autoregressive coefficients satisfy the stationarity condition, the process with initial values converges to the stationary process in the long run.

Define an autoregressive process with given past values.

In[1]:=
Click for copyable input
X
Out[1]=

Find the mean function of the process.

In[2]:=
Click for copyable input
X
Out[2]=

Form conditions for longrun stability of the mean function.

In[3]:=
Click for copyable input
X
Out[3]=

Verify that these conditions are equivalent to weak stationarity condition.

In[4]:=
Click for copyable input
X
Out[4]=
In[5]:=
Click for copyable input
X
Out[5]=

Visualize the parameter region and three sample parameter pointsinside, outside, and on the boundary of the region of weak stationarity.

In[6]:=
Click for copyable input
X
Out[6]=

Visualize sample trajectories for a few parameter choices.

In[7]:=
Click for copyable input
X
Out[7]=
In[8]:=
Click for copyable input
X
Out[8]=

Visualize mean functions for these parameter choices.

In[9]:=
Click for copyable input
X
Out[9]=