![]() Future years are mostly Statista projections These projections or forecasts are conducted by regression analyses, exponential trend smoothing (ETS) or similar techniques and extrapolate the found historical trend. As new data becomes available or methodologies are adapted to suit changing requirements it can be possible that data is not comparable any longer with previously published data or is changed retroactively according to the new definitions.īecause of the high degree of processing no specific external source can be named for each data point and all data for historical years (usually until the last finished year before the current one) have to be considered Statista estimates. In Croatia, India, Singapore and Sweden, the recommended limit is 10 grams of alcohol (less than. In 2017, 20 of the population reported not drinking at all 2 and overall consumption has fallen by around 16 since 2004 3. Most indicators are composites of multiple input sources with slightly varying methodologies that have been processed by our analysts to be aligned and consistent with each other and with all other indicators in the KMI database. The countries also had very different recommendations for the limits of daily and weekly drinking. Data for missing countries or regions are imputed by considering known information from other countries or regions that are found to be similar by cluster analyses like k-means or similar procedures. Measured through sales data, overall alcohol consumption averaged 8.7 litres per person across OECD countries in 2019, down from 9.1 litres in 2009 (Figure 4.3). Data for missing years are interpolated by various statistical means, such as linear or exponential interpolation or cubic splines. These datasets are often incomplete as there are gaps between survey years or no or no reliable information might be available for a specific indicator in a specific country or region. ![]() (per capita) and nearly three times as much as the worldwide average (6.3 liters of pure alcohol per person over the age of 15 over the course of one year). Whereas primary data are generated via Statista's own surveys like the Global Consumer Survey, secondary input datasets are mostly sourced from international institutions (such as the IMF, the World Bank or the United Nations), national statistical offices, trade associations and from the trade press. Situated between Romania and Ukraine, this former Soviet republic consumes just about twice as much alcohol as the U.S. The shown forecasts represent a blend of multiple input datasets from both internal (primary) and external (secondary) sources. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |