Documente Academic
Documente Profesional
Documente Cultură
sezonalitate
Media mobilă dublă
(double moving average)
Method Theil's U
Double Moving Average 0.1779
Method Parameter Value
Double Moving Average Order 4
Double Exponential Smoothing
reprezentare grafică
Double Exponential Smoothing
(Initialization) L1 = Y1, T1 = 0, F1=Y1
Level: Lt = α * Yt + (1 – α) * (Lt-1 + Tt-1)
Trend: Tt = β * (Lt – Lt-1) + (1 – β) * Tt-1
Fit: Ft = α * Yt-1 + (1 – α) * Ft-1
Forecast for period m: Ft+m = Lt + m*Tt
Note:
First fit is available from period 2.
Double Exponential Smoothing
output Oracle Crystal Ball
Period Lower: 2.5% Forecast Upper: 97.5%
13 31.48 33.37 35.26
14 32.29 35.45 38.61
15 32.52 37.53 42.55
16 32.56 39.61 46.66
Statistic Historical data
Ljung-Box 11.88
Seasonality Non-seasonal
Method Rank RMSE
Double Exponential Smoothing Best 0.96
Method Theil's U
Double Exponential Smoothing 0.5857
Method Parameter Value
Double Exponential Smoothing Alpha 0.6631
Beta 0.9990
Aplicaţie
Determinaţi ajustările pentru anii 5 şi 6 prin metoda
nivelării exponenţiale duble
Dacă cifra de afaceri estimată pentru anul 12 este 28,2 să
se realizeze previziunile pentru anii 13 şi 14
Damped Trend Smoothing
Nonseasonal
reprezentare grafică
Damped Trend Non-Seasonal
output Oracle Crystal Ball
Period Lower Forecast Upper
13 31.47 33.36 35.25
14 32.28 35.44 38.60
15 32.50 37.52 42.53
16 32.55 39.59 46.63
Statistic Historical data
Ljung-Box 11.88
Seasonality Non-seasonal
Method Rank RMSE
Damped Trend Non-Seasonal Best 0.96