Example analyses

Airline passenger volume
A classic data set demonstrating non-stationary periodicity
Solar irradiance
The automatic statistician identifies the unusual period of very low sun spot activity known as the Maunder minimum
Sulphuric acid production
Creative usage of change points
Unemployment figures
An example where the automatic statistician identifies that it has failed to capture all aspects of the data (see the model checking section)
Number of affairs
A simple but insightful model. Gaussian noise clearly inappropriate but identified by appropriate model criticism.
Stove smoke
A simple analysis of data relating to wood burning stoves. Only a linear model is required to find patterns in the data that would not be apparent from looking at plots of pairs of variables alone.
“It would be very nice to have a formal apparatus that gives us some ‘optimal’ way of recognizing unusual phenomena and inventing new classes of hypotheses that are most likely to contain the true one; but this remains an art for the creative human mind.” E. T. Jaynes 1985