Model Output Statistics (MOS) is an objective weather forecasting technique that determines a statistical relationship between a predictand and variables forecast by a numerical model. It has been applied to predict surface wind, probability of precipitation, maximum temperature, cloud amount, and conditional probability of frozen precipitation. Predictors include surface observations and model outputs from the Subsynoptic Advection Model (SAM) and the Primitive Equation (PE) model. Verification scores have been computed and compared to other objective techniques and official forecasts. MOS forecasts are disseminated by the National Weather Service (NWS) via teletype and facsimile. MOS combines dynamical and statistical methods for forecasting. It involves screening regression, a technique that selects predictors to include in a regression equation. This method has been used to develop equations for forecasting various weather variables. For example, MOS has been used to estimate probability of precipitation (PoP), with equations developed for different time periods and stations. These equations have shown better performance than local forecasts in some cases. Surface wind forecasts using MOS have been developed for 10 stations in the eastern United States. Equations for wind components and speed were derived using SAM and PE model outputs. These equations have been compared to official forecasts and have shown good performance in terms of direction and speed. Maximum temperature forecasts using MOS have been developed for 16 stations. These equations include predictors such as temperature, humidity, and cloud cover. The equations have been compared to other forecasting methods and have shown good performance. Cloud amount forecasts have been developed for four stations using regression equations. These equations have shown good performance in terms of accuracy. Conditional probability of frozen precipitation (PoFP) has been estimated using a subsample of data where precipitation occurred. The main predictor is the Wagner index, which is a conditional probability of frozen precipitation based on the 1000–500 mb thickness. These forecasts have shown good reliability. MOS forecasts are transmitted twice daily by the NWS via teletype and facsimile. These forecasts are used in experimental, computer-produced worded forecasts. MOS has been shown to be a useful technique in objective weather forecasting, particularly for probabilistic forecasts. The technique has been applied to various weather variables and has shown good performance compared to other methods.Model Output Statistics (MOS) is an objective weather forecasting technique that determines a statistical relationship between a predictand and variables forecast by a numerical model. It has been applied to predict surface wind, probability of precipitation, maximum temperature, cloud amount, and conditional probability of frozen precipitation. Predictors include surface observations and model outputs from the Subsynoptic Advection Model (SAM) and the Primitive Equation (PE) model. Verification scores have been computed and compared to other objective techniques and official forecasts. MOS forecasts are disseminated by the National Weather Service (NWS) via teletype and facsimile. MOS combines dynamical and statistical methods for forecasting. It involves screening regression, a technique that selects predictors to include in a regression equation. This method has been used to develop equations for forecasting various weather variables. For example, MOS has been used to estimate probability of precipitation (PoP), with equations developed for different time periods and stations. These equations have shown better performance than local forecasts in some cases. Surface wind forecasts using MOS have been developed for 10 stations in the eastern United States. Equations for wind components and speed were derived using SAM and PE model outputs. These equations have been compared to official forecasts and have shown good performance in terms of direction and speed. Maximum temperature forecasts using MOS have been developed for 16 stations. These equations include predictors such as temperature, humidity, and cloud cover. The equations have been compared to other forecasting methods and have shown good performance. Cloud amount forecasts have been developed for four stations using regression equations. These equations have shown good performance in terms of accuracy. Conditional probability of frozen precipitation (PoFP) has been estimated using a subsample of data where precipitation occurred. The main predictor is the Wagner index, which is a conditional probability of frozen precipitation based on the 1000–500 mb thickness. These forecasts have shown good reliability. MOS forecasts are transmitted twice daily by the NWS via teletype and facsimile. These forecasts are used in experimental, computer-produced worded forecasts. MOS has been shown to be a useful technique in objective weather forecasting, particularly for probabilistic forecasts. The technique has been applied to various weather variables and has shown good performance compared to other methods.