step of forecasting

1. Steps Of Forecasting Group 2

2. Determine the use of the forecast Who needs the forecast? All organizations operate in the atmosphere of uncertainty. Decisions to be made affects future of the organization.

3. Select the items to be forecasted The item to be forecasted. Dependent variable to be studied.

4. Determine the time horizon of the forecast Short-range forecast Up to 1 year Purchasing, job scheduling, job assignments Medium-range forecast 1 year to 3 years Sales and production planning Long-range forecast 3 + years New product planning, research and development

5. Select Forecasting approach Qualitative Methods Used when situation is vague and little data exist New products New technology Involves intuition, experience

6. Quantitative Methods Used when situation is ‘stable’ and historical data exist Existing products Current technology Involves mathematical techniques

7. Data collection One of the most difficult and time consuming part of forecasting is the collection of valid and reliable data. Forecast can be no more accurate than the data on which it is based Data can be collected from- primary source and secondary source

8. Four criteria can be applied to the determination of whether the data will be useful- Data should be reliable and accurate Data should be relevant Data should be consistent Data should be timely

9. Data Reduction Since available data can be either too much or too less, data reduction is necessary. Decide which data is most complete, valid and reliable to increase data accuracy. Some times accurate data may be available but only in certain historic periods.

10. Exploring Time Series Data Patterns Horizontal pattern- When data observation fluctuate around a constant level or mean Trend pattern- When data observation grow or decline over an extended period of time Cyclic pattern- When data observation exhibits rises and falls that are not of a fixed period Seasonal Pattern- When data observation are influenced by seasonal factors.

11. Exploring Data Patterns with Auto correlation Analysis Autocorrelation is the correlation between a variable lagged one or more period itself. It is used to detect non randomness of data To identify an appropriate time series model if data is not random

12. Time t Month Original data Y lagged one period Y lagged 2 period 1 January 123     2 February 130 123   3 March 125 130 123 4 April 138 125 130 5 May 145 138 125 6 June 142 145 138 7 July 141 142 145 8 August 146 141 142 9 September 147 146 141 10 October 157 147 146 11 November 150 157 147 12 December 160 150 157

13. Time t Yt Yt-1 Yt - Y Yt-1 - Y   ( Yt - Y   ) ^2 ( Yt – Y) ( Yt-1 - Y )   1 123   -19   361   2 130 123 -12 -19 144 228 3 125 130 -17 -12 289 204 4 138 125 -4 -17 16 68 5 145 138 3 -4 9 -12 6 142 145 3 7 141 142 -1 1 8 146 141 4 -1 16 -4 9 147 146 5 4 25 20 10 157 147 15 5 225 75 11 150 157 8 15 64 120 12 160 150 18 8 324 144   Total 1704     1474 843

14. Y= 1704/12 = 142 r 1 = 843/1474 = .572

15. Select the forecasting model(s) The most prominently used models are: Exponential smoothing method with 1 or 2 variables. Regression Models Once the model has been judicially selected, its parameters are estimated for model fitting purposes.

16. Make the forecast Forecast is made for a particular period.

17. Forecast evaluation Comparing Forecast value with actual historical values. ……………………… ^ Error : e t = y t –y t

18. Thank you