ADVANCED WEATHER VERIFICATION SCORING ALGORITHMS 2
In this second series, we provide a detailed description of the weather scoring algorithms currently incorporated into the Mine-Weather system. It uses a number of scoring models based on statistical methods to verify weather attributes (temperature, precipitation, wind speed/gust, etc.) in the form of scores. When considering numeric variables (point forecasts like temperature, % forecast Humidity, etc.) the scoring is based on the numeric comparisons of forecasted values against the observed actuals. For dichotomous attributes, we compare the forecasts against actuals by using statistical contingency tables as the basis for the score calculations. Mine-Weather also has capabilities for verifying special Weather Events (hurricanes, winter storms, etc.) that are of great importance to utilities.
In this second series, we provide a detailed description of the weather scoring algorithms currently incorporated into the Mine-Weather system. It uses a number of scoring models based on statistical methods to verify weather attributes (temperature, precipitation, wind speed/gust, etc.) in the form of scores. When considering numeric variables (point forecasts like temperature, % forecast Humidity, etc.) the scoring is based on the numeric comparisons of forecasted values against the observed actuals. For dichotomous attributes, we compare the forecasts against actuals by using statistical contingency tables as the basis for the score calculations. Mine-Weather also has capabilities for verifying special Weather Events (hurricanes, winter storms, etc.) that are of great importance to utilities.
Download this paper to learn more about the second series of advanced Weather Verification Scoring Algorithms.
