Remote Sensing for Short-Term Economic Forecasts

Economic forecasts are an important instrument to judge the nation-wide economic situation. Such mainly based on data from statistical offices. However, there is a time lag between end of reporting period and release that arises for instance needed collect process data. To improve by reducing delay, it interest find alternative sources provide information activity without significant delays. Among others, satellite images thought assist here. This paper addresses potential earth observation imagery short-term forecasts. The study focused estimation investments in construction sector high resolution (HR) (10–20 m) very (VHR) (0.3–0.5 as well agricultural machinery orthophotos (0.1 simulating VHR imagery. By applying machine learning possible extract objects certain extent. For detection areas, much better suited than HR images. with ground 30–50 cm able identify machinery. These results promising new unconventional input forecasting models.