Forecasting method MAPE MAD MSD Multiplicative decomposition 8.35 67.66 7573.78 model with trend and seasonality Additive decomposition model 8.16 65.93 7467.41 with trend and seasonality Moving average 14.2 128.9 33243.4 Single exponential smoothing
6.5 53.6 15457 Holt's method 13.2 110.6 27050.4 Trend analysis 15.39 134.16 32031.66 Winters multiplicative model 5.38 43.92 5089.33 Winters additive model 5.06 40.58 4705.68 Figure 2: Comparison of actual sales with forecasted demand in SMA method.
Olsen and Jose (1982) used data on two restaurants to compare single and double exponential smoothing
Use the single exponential smoothing
method to predict the Hurst coefficient in future time intervals.
The first is considered to be simple and refers to the exponential smoothing
models whereas the second characterizes the Box-Jenkins methodology.
The combination of autoregressive integrated moving average (ARIMA) and seasonal exponential smoothing
(SES) techniques were found to be the most accurate.
Also the authors of this study would like to implement weighting method to weight the three criteria based on exponential smoothing
for multi-criteria inventory classification (Jamshidi & Jain, 2008).
To improve the results in the event that excessive variation in the electricity consumption during the occupied or unoccupied hours causes the first and last knot points to incorrectly identify the startup and shutdown times, exponential smoothing
with adaptive updating of the discount factor can be used to smooth the hourly observations before solving the constrained optimization problem in Equation 2.
Let be a popularity degree of a block is defined as the future access frequency based on the number of access demand, at a time, the popularity degree of a block can be calculated using Holt's Linear and Exponential Smoothing
(2006) Exponential Smoothing
Model Selection for Fore-Casting.
Smoothers (and parameters in italics) used are lowess (weighted quadratic least-squares), f (span); Friedman super smoother (cross-validated running lines), span (span); cubic spline, spar (smoothing parameter), Savitzky-Golay (local polynomial regression), p, f (polynomial order, frame size); Kalman (state space model), none; running mean, wsize (window size); Whittaker (recursive penalized B-splines), [lambda] (spline parameter); and exponential moving average (EMA) (weighted recursive exponential smoothing
), [alpha] (spline parameter).
In Pakistan there is a need to stabilize production of wheat in order to sustain food security the main crop models used for wheat in our country are APSIM DSSAT SWAT CROPWAT Random Walk with Drift Linear Trend Simple Exponential Smoothing
Whether you know R programming language or not, Apache Mahout or concepts such as Holt-Winters multiplicative exponential smoothing
: data mining will be sprinkled into a lot of modern applications in clever ways, hiding all the complexities but keeping the magic.