In order to determine how well the AP macroeconomic models do at predicting the future, one needs to know what the model is actually looking at. The models are designed in such a way that they take the data and make predictions about the economy in the future. They then try to explain why the data was produced the way it was. To see how good the model is you need to understand what they are trying to predict.
AP models were developed by the American Enterprise Institute as part of their macroeconomic forecasting project which has been going on since the 1970s. The data they use is the kind of information you would normally expect to find in any normal economic forecasting publication or research project.
The models will try to provide an explanation for why the data was produced in such a way that they can predict the future economic activity. For instance, if there was an outbreak of the bubonic plague in some countries, they might argue that the disease was brought on by poor sanitation. They then claim that they know why this is the case and how to prevent it in the future.
This is all part of AP's macroeconomic forecasting process and they also claim that they can predict where the next outbreak of this disease will occur. They use the same methodology that they used to predict the outbreak of the global financial crisis. Unfortunately, this approach was not correct and the spread of the disease was far faster than the models predicted. It is important to note that the model that was used by the AP did not predict the severity of the current economic crisis.
Many people do not fully understand the concepts behind AP's models. Some of the assumptions that are included in their models are that inflation will stay above certain levels and that interest rates will remain at their current level. In some cases they also assume that the central bank is stable and does not have a plan to intervene and change interest rates and inflation.
These assumptions are made in order to help the AP's models generate accurate predictions. However, they are important to understand because they do not allow the models to be tested on the real world. They will not be able to test their predictions on events that are likely to happen within the next few months and years.