Population Regression Model

The underlying population regression model, motivated by the preceding theoretical expectations, isn’t explicitly specified by Braithwaite but can be easily specified as follows:

$$Spread = \beta_0 + \beta_1 Territory+\beta_2 Resources+\beta_3Mountain+\beta_4 Forrest+ \beta_5Ocean$$

$$+ \beta_6 Size\ of\ States+\beta_7 VitalBorder+\beta_8 Control.variable+ \epsilon$$

According to Braithwaite, the dependent variable, $spread$, is a circular area measured in squared kilometers, log transformed. The key independent and control variables are on the right-hand side, with all the $\beta_s$ indicating the marginal effects of respective right-hand side variables. Drawn from the article, the key independent variables, their measurements, and expected effects are as follows in the table below.