Smoothing Forecast Analysis

Week | Widgets | Smoothed | |

No. | Sold | Forecast | |

1 | 40 | #N/A | |

2 | 90 | 40.00 | |

3 | 100 | 80.00 | |

4 | 110 | 96.00 | |

5 | 90 | 107.20 | |

6 | 120 | 93.44 | |

7 | 109 | 114.69 | |

8 | 96 | 110.14 | |

9 | 80 | 98.83 | |

10 | 120 | 83.77 | |

11 | 130 | 112.75 | |

12 | 200 | 126.55 | |

13 | 190 | 185.31 | |

14 | 180 | 189.06 | |

15 | 200 | 181.81 | |

Forecast For Week 16 | 196.36 | ||

The fundamental
idea behind smoothing is that each new forecast is obtained in part by moving the prior
forecast in a direction that would have improved the old forecast. The equation is: F[t+1]
= F[t] + a x e[t] where: 1. t is the time period 2. F[t] is the forecast at time t, and F[t+1] is the forecast at the time period immediately following time t 3. a is the smoothing constant (Never exceed .7) 4. e[t] is the error: the difference between the forecast for time t and the actual observation at time t How do you use it? Enter your business units (widgets) and this sheet will forecast unit sales for you. |
|||

© Copyright, 2014, JaxWorks, All Rights Reserved |