Exponential smoothing and Sales forecasting software.
Sales Forecasting software uses statistical forecasting methods to generate forecasting. We at MXI Software explain relevance of exponential smoothing for sales forecasting, budgeting and business modelling.
Exponential smoothing methods weight the historical data using exponentially decreasing weighting. The immeidiate prior period has the most weight and each period prior to it has relatively less weight. The decline in weight is expressed mathematically as an exponential function. The smoothing parameters determine the weights. To see the relevance of this in action we developed the mxi free forecasting offer. MXI 60 minute free forecast session:
Comparison Among Exponential Smoothing Methods
Single Exponential Smoothing: Identifies the percentage of weight given to the prior period and all other historical periods. It does not adjust for trend or for seasonal variance.
Double Exponential Smoothing: Finds trend then adjusts the forecast data to reflect this trend instead of generating a single parameter for all forecast periods.
Holt-Winters: Identifies both trend and seasonal variance, and adjusts the forecast data to reflect these factors. This method is tuned to both high and low outliers. A better choice for handling seasonality is Double Exponential Smoothing with the Data Filters parameter set to Seasonal Adjustment.
Advanced Parameters for Exponential Smoothing
These smoothing constants are used in the equations for exponential smoothing methods. Keep the default settings unless you have a strong background in time-series forecasting.
Alpha: Determines how responsive a forecast is to sudden jumps and drops. It is the percentage weight given to the prior period, and the remainder is distributed to the other historical periods. Alpha is used in all exponential smoothing methods.
The lower the value of alpha, the less responsive the forecast is to sudden change. A value of 0.5 is very responsive. A value of 1.0 gives 100% of the weight to the prior period, and gives the same results as a prior period calculation. A value of 0.0 eliminates the prior period from the analysis.
Beta: Determines how sensitive a forecast is to the trend. The smaller the value of beta, the less weight is given to the trend. The value of beta is usually small, because trend is a long-term effect. Beta is not used in Single Exponential Smoothing.
Gamma: Determines how sensitive a forecast is to seasonal factors. The smaller the value of gamma, the less weight is given to seasonal factors. Gamma is used only by the Holt-Winters method.
Trend Dampening: Determines how sensitive the forecast is to large trends in recent time periods. Dampening identifies how quickly the trend reverts to the mean. A higher value implies slower dampening while a lower value implies faster dampening. The smaller the value, the less effect the trend has on the forecast.
For each constant, you can specify a maximum value, a minimum value, and an interval. The interval is an incremental value between the maximum and minimum, which the forecasting engine uses to find the optimal value of the constant.
Contact MXI Software