Forecasting with Mixed Data Sampling Models (MIDAS) and Google trends data: the case of car sales in Italy

In the last few years the attention focused on the use of Internet data as an information source that could improve forecasts. Google Trends data can provide weekly information that represent proxies for market expectations of the current economic fundamentals and, therefore, may be helpful for short-term economic predictions. Unfortunately, most economic activity indicators are sampled at a lower frequency than the week, therefore the two datasets are unbalanced. In the present paper we use a Mixed Data Sampling Model to forecast monthly time series with weekly real-time information obtained from Google trends. The model is applied to car sales in Italy