Vector Joint Model vs. Univariate Component Models 

Get hourly temperature readings for May 2014 in Champaign, Illinois.

In[1]:=
Click for copyable input
X
In[2]:=
Click for copyable input
X

Use TimeSeriesAggregate to calculate daily minimum and maximum temperatures.

In[3]:=
Click for copyable input
X

Combine them into a vector time series.

In[4]:=
Click for copyable input
X
Out[4]=
show complete Wolfram Language input
Out[5]=

The first part of the data will be used to find a model, while the rest of the data will serve as a reference set for the forecast.

In[6]:=
Click for copyable input
X

The temperatures are cross-correlated.

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

Fit a vector model to the data.

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

Forecast the next 5 days.

In[9]:=
Click for copyable input
X
show complete Wolfram Language input
Out[10]=

Find univariate models of the same type and but longer orders for both temperatures separately.

In[11]:=
Click for copyable input
X
Out[11]=
In[12]:=
Click for copyable input
X
Out[12]=

Combine univariate forecasts for plotting.

In[13]:=
Click for copyable input
X
Out[13]=

Compare the forecasts.

In[14]:=
Click for copyable input
X
show complete Wolfram Language input
Out[15]=

Plot the vector forecast and the 95% confidence bands.

show complete Wolfram Language input
Out[17]=

Plot the univariate forecasts and the 95% confidence bands.

show complete Wolfram Language input
Out[20]=

Compare both forecasts and the corresponding confidence bands.

show complete Wolfram Language input
Out[21]=
es