Western Europe in the Middle Ages: Demographic Transition
In this posting, I look at the Demographic Transition and Malthusian controller in Medieval Western Europe. Typically, these two controllers are not thought to do together.
The best short-term (year-to-year) model had Technical Productivity (TECHP) as an input; the model was nonlinear, unstable and cyclical. The best Attractor Path model as the BAU model which was cyclical and unstable (see below). All the AIC statistics (except the W_Input model) had overlapping confidence intervals and were equally good competitors.
The WE_MA_BAU System matrix was unstable (in two components, WE1 and WE2) and cyclical. It is actualy very close to a Random Walk. See the WE_MA BAU Model on my Google Site here for instructions on modifying the BAU model to create an RW model.
Notes
Galor, Oded (2011). Unified Growth Theory. Princeton: Princeton University Press.
Boulding, K. (1955) The Malthusian Model as a General System
The Malthusian Controller (Q-N) System Theoretic interpretation of the Malthusian Model.
For more information about data sources how the state space models were estimated see the Boiler Plate.
You can run the WE_MA BAU Model on my Google Site here. Instructions in the R-code explain how to modify the model.
You can run the W_MA BAU Model on my Google Site here. Instructions in the R-code explain how to modify the model.
Wikipedia
WE_MA Measurement Model
The Western European Middle Ages WE_MA BAU Model has three component state variables that explain 100% of the variation in the indicators. The first component is WE1 = (Overall Growth). The second component is WE2 = (Q-U) an Urban-Production Controller. The third component is WE3 = (N-U-Q) an Urban-Malthusian-Population controller.
WE_MA AIC Statistics
The best short-term (year-to-year) model had Technical Productivity (TECHP) as an input; the model was nonlinear, unstable and cyclical. The best Attractor Path model as the BAU model which was cyclical and unstable (see below). All the AIC statistics (except the W_Input model) had overlapping confidence intervals and were equally good competitors.
WE_MA System Matrix
The WE_MA_BAU System matrix was unstable (in two components, WE1 and WE2) and cyclical. It is actualy very close to a Random Walk. See the WE_MA BAU Model on my Google Site here for instructions on modifying the BAU model to create an RW model.
Comments
Post a Comment