forecasting
DESCRIPTION
ForecastingTRANSCRIPT
Short Term Forcasting
Month cummluative avg
Jan 1 -Feb 1.3 1.00 - W1=0.1, W2=0.3, W3=0.6Mar 2.7 1.15 -Apr 3.1 1.67 1.67 2.11May 3.8 2.03 2.37 2.8Jun 4.1 2.38 3.20 3.48Jul 2.67 3.67 3.91
USED Stable environment Used Most Frequently as weight can Moderated Better Forcast
Sum of all Weights equal to 1
Weights may be subjectively ae.g. W1 W2 W31,2,3,4=10(sum)
W1= 1/10W2= 2/10W3= 3/10W4=4/10
The largest weight should be
Demand in(000)
n- point moving avg (3 month moving avg)
n- point weighted moving avg (3 month weighthed
moving avg)
Short Term Forcasting
1.001.001.252.442.983.654.02
New Prodcut LaunchesBetter Forcast
F(t+1)=Forecast next time periodA(t)= Actual of this time periodF(t)=Forcast of the time period
the forcast of the first time period may be taken anyone of the foloowing three ways 1. it may be asssumed or given2. it may be taken as actual of the first time period3. it may be taken to be avg of the first time periods
Single exponential smoothingSES =α= 0.82
Single exponential smoothingSES =α= 0.82
F(t+1) = αA(t)+(1- α)F(t)
Time Series Linear Regression
x Month SUMMARY OUTPUT
1 Jan 12 Feb 1.3 Regression Statistics3 Mar 2.7 Multiple R 0.97828914 Apr 3.1 R Square 0.95704965 May 3.8 Adjusted R 0.946312
6 Jun 4.1Standard Er0.2962464
7 JulObservatio 6
a(intecept) 0.3266667 ANOVAb (slope) 0.6685714 df SS MS
r (Corelation) 0.9782891 Regression 1 7.8222857 7.8222857R Square 0.9570496 Residual 4 0.3510476 0.0877619
Forecast for July 5.0066667 Total 5 8.1733333
CoefficientsStandard Error t StatIntercept 0.3266667 0.2757904 1.1844746x 0.6685714 0.0708164 9.4409091
Demand in(000)
Time Seriesy=a+bx
y= Dependent Variable
F Significance F89.130765 0.0007019 y=.32+.668x
P-value Lower 95% Upper 95%Lower 95.0%Upper 95.0%0.3017961 -0.43905 1.0923834 -0.43905 1.09238340.0007019 0.4719535 0.8651893 0.4719535 0.8651893
were a is Constant or Intercept
b= is slope/Radiant /Regression Coefficent
x= independent variable/ predict variabel /treatment variable