Economic stability is crucial for creating a predictable environment needed for growth. Economic activity is commonly measured by GDP. Although GDP is not an ideal indicator of economic stability, we find that it reflects the market side of the economy in the long term. Furthermore, we find that GDP trends have real implications for economic activity.
GDP is modelled as the sum of its expenditure components - private consumption, government spending, gross investment and net exports. Due to stationarity issues, we calculate growth rates of GDP components instead of working with monetary data, as working with non-stationary variables can lead to unreliable conclusions about correlations and causality of time series.
We model GDP using a vector autoregressive model that allows us to model each GDP component as a linear combination of each GDP components' own lagged values as well as lagged values of other components. Compared to univariate models, VAR models allow us more flexibility that is needed for modelling interconnected time series.
The best VAR model is achieved by smoothing crisis periods in our data, that complicate model assessment and do not reflect usual GDP trends. Slovenia's economy reacts strongly to shocks, but shocks fade quickly in most cases. The estimated responses of some components to shocks are not reliable due to wide confidence intervals. In the future, we expect long-term growth in Slovenia's GDP due to increased private consumption, government spending, and gross investment, but net exports are expected to decline.
|
