3.4.2. Estimated Regression Model
Given that all variables are stationary at levels, we have opted to use the ordinary least squares (OLS) estimation
method. It appears that the estimated model (see table 2) is a good fit given that about 99 % of the variability in the
dependent variable is fully captured in the model. The estimated model diagnostics seem to satisfy a priori
econometric test (see appendices A, B, and C). Our results hint that autonomous consumption (constant) is positive,
but it is statistically insignificant.
The estimated model provides robust evidence that household spending is positively associated with crosscultural
dynamics (proxied by social globalization index); a unit rise in social globalization index spurs domestic
household spending by 1.7 % (statistically significant at the 0.05 level). Further evidence provides support that
household spending is positively associated with net disposable household income; this implies that a 1 % increase
in net disposable income accelerates spending by 0.88 % (statistically significant at the 0.01 level).