Monday, December 31, 2007

How well does real GDP growth predict change in GDP per capita in PPS?

In one of my previous posts, I predicted that by 2009, Slovakia will overtake Portugal in GDP per capita in PPS. I calculated this from from real GDP growth forecasts of the European Commission (Eurostat) for 2009 and the 2008 forecasts of GDP per capita in PPS. How accurate is however this prediction?

For evaluating this, I graphed above the actual annual changes in GDP per capita in PPS relative to the EU27 average (x-axis; in percentage points) and the values predicted via calculation of real GDP growth (y-axis), using a very small dataset (n=13) of Slovakia between 1996-2008 (Data source: Eurostat1, Eurostat2). The correlation coefficient is 0.9304, the trend line is defined as y = 1.0038x - 0.2169 (R=1 would mean y = 1x + 0). The standard deviation of the actual values from the predictive model is however still too high for a conclusive answer about a minute positive difference between Slovakia and Portugal in 2009.

No comments: