1 forecasting models and methods dr. geurts lecture

6
1 FORECASTING MODELS AND METHODS Dr. Geurts Lecture

Upload: curtis-mason

Post on 19-Jan-2016

212 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: 1 FORECASTING MODELS AND METHODS Dr. Geurts Lecture

1

FORECASTING MODELS AND METHODS

Dr. Geurts Lecture

Page 2: 1 FORECASTING MODELS AND METHODS Dr. Geurts Lecture

2

New Products 1. Buyer intention surveys:

Definitely will buy Highly possible As likely as unlikely to buy highly doubtful to buy Definitely will not buy

2. Test markets 3. Flow through or economic study 4. Diffusion models = rate of adoption imitators/innovators 5. Product comparable

Page 3: 1 FORECASTING MODELS AND METHODS Dr. Geurts Lecture

3

New Products- con’t

6. Judgmental models = expert guess, Delphi 7. New product models like dumps:

Durability Number of potential Users Number of Major competitors Number of Potential customers Proportion made aware [ MARKET Share ]

8. Conjoint analysis = determine value of product attributes an estimate market share.

9. Trend/fashion forecasts = Innovators

Page 4: 1 FORECASTING MODELS AND METHODS Dr. Geurts Lecture

4

Existing Product Sales 10. Business activity = capacity being used 11. Time series = past data patterns

a. exponential smoothing b. Box-Jenkins [ARIMA]

12. Response models = Sales and marketing mix variables 13. Econometric models = Using economic indicators 14. Salesmen composite = Summation of salesman forecast

for her territory 15. Logistic regression = combination of marketing mix

variables and time series forecasts.

Page 5: 1 FORECASTING MODELS AND METHODS Dr. Geurts Lecture

5

Other Forecasting Methods

15. Technological 16. Combining forecasts 17. Partitioned data 18. Regression 19. Interest rates 20. Economic Growth

Page 6: 1 FORECASTING MODELS AND METHODS Dr. Geurts Lecture

6

Role of Data Bad Past Other forecasts

Uses of Forecasts Budgets Production quantities Inventory control Planning Bank loans Identify effect of problems or promotions. If the forecasting has been

accurate and the company runs a promotion. The company can measure the effects of the promotion as the difference between forecast and actual.