| dc.description.abstract | This Master thesis investigates nowcasting, or predicting real time GDP, power of the
following series: i) hard data gauging real economy; ii) soft indicators reflecting business and
financial markets’ sentiment and iii) uncertainty measures depicting the overall uncertainty in
Norway. I employ approximate dynamic factor model, a framework acknowledged by
researchers and practitioners at central banks, to examine the predictive power of 209 variables
sorted in 15 blocks according to their economic content and release time. This thesis
documents that finance related variables are good in predicting the current state of the
economy. Due to their timely release and forward looking nature, they also perform well in
forecasting the economic growth over the following year. These findings suggest that finance
related variables are useful inputs for conducting timely and adequate monetary policy.
Uncertainty measures help to predict the contemporaneous economic growth rate as well. Real
variables like industrial production, while released with a lag and at a later date, add to the
precision of the nowcast the most. | nb_NO |
