Anúncio
Anúncio

Mais conteúdo relacionado

Anúncio

forecasting.ppt

  1. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-1 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Quantitative Analysis for Management Forecasting
  2. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-2 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Forecasting Models Moving Average Exponential Smoothing Trend Projections Time Series Methods Forecasting Techniques Delphi Methods Jury of Executive Opinion Sales Force Composite Consumer Market Survey Qualitative Models Causal Methods Regression Analysis Multiple Regression
  3. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-3 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Scatter Diagram for Sales 0 50 100 150 200 250 300 350 400 450 0 2 4 6 8 10 12 Time (Years) Annual Sales Televisions
  4. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-4 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Decomposition of Time Series Time series can be decomposed into: Trend (T): gradual up or down movement over time Seasonality (S): pattern of fluctuations above or below trend line that occurs every year Cycles(C): patterns in data that occur every several years Random variations (R): “blips”in the data caused by chance and unusual situations
  5. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-5 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Product Demand Showing Components -150 -50 50 150 250 350 450 550 650 0 1 2 3 4 5 Time (Years Demand Trend Actual Data Cyclic Random
  6. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-6 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Moving Averages n n) period in (demand : average Moving 
  7. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-7 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Calculation of Three-Month Moving Average Month Actual Shed Sales Three-Month Moving Average January 10 February 12 March 13 April 16 3 2 11 13)/3 12 (10 = + + May 19 3 2 13 16)/3 13 (12 = + + June 23 16 19)/3 16 (13 = + + July 26 3 1 19 23)/3 19 (16 = + +
  8. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-8 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Weighted Moving Averages weights ) period in )(demand period for (weight average moving Weighted   = n n
  9. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-9 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Calculating Weighted Moving Averages Weights Applied Period 3 Last month 2 Two months ago 1 Three months ago 3*Sales last month + 2*Sales two months ago + 1*Sales three months ago 6 Sum of weights
  10. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-10 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Calculation of Three-Month Moving Average Month Actual Shed Sales Three-Month Moving Average January 10 February 12 March 13 April 16 6 1 12 10)]/6 * (1 12) * (2 13) * [(3 = + + May 19 3 1 14 12)]/6 * (1 13) * (2 16) * [(3 = + + June 23 17 13)]/6 * (1 16) * (2 19) * [(3 = + + July 26 2 1 20 16)]/6 * (1 19) * (2 23) * [(3 = + +
  11. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-11 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Exponential Smoothing New forecast = previous forecast + (previous actual - previous) or: where ( ) 1 1 1  - - - - + = t t t t F A F F actual period previous constant between 0~1 smoothing forecast previous forecast new = = = = -1 1  t t- t A F F
  12. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-12 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Table 5.5
  13. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-13 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Table 5.5 Continued  =0.50 Qtr Actual Tonnage Unloaded Forecast using  =0.50 1 180 175 2 168 177.50 =175.00+0.50(180-175) 3 159 172.75 =177.50+0.50(168-177.50) 4 175 165.38 =172.25+0.50(159-172.25) 5 190 170.19 =165.38+0.50(175-165.38) 6 205 179.09 =170.19+0.50(190-170.19) 7 180 179.54 =179.09+0.50(180-179.09) 8 182 182.00 =179.54+0.50(182-179.54) 9 ?
  14. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-14 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Trend Projection General regression equation: + = 2 X 2 n X Y X n XY b Y a Y where bX a Y - - = = =   intercept axis - variable) (dependent predicted be to variable the of value computed
  15. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-15 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Table5.7
  16. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-16 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Solved Formula
  17. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-17 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Midwestern Manufacturing’s Demand 60 70 80 90 100 110 120 130 140 150 160 1993 1994 1995 1996 1997 1998 1999 2000 2001 Forecast points Trend Line Actual demand line X y 54 . 10 70 . 56 + =
  18. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-18 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Computing Seasonality Indices Using Answering Machine Sales
  19. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-19 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Trend Analysis with Seasonal Indices Y = 1150 + 20x Where x=1,2,…12 for Jan, Feb,….Dec So; Jan =[1150+20(1)]*.957 = 1119.69 Feb =[1150+50(2)]*.851 = 1012.69 Mar =[1150+20(3)]*.904 = 1093.84 . . Dec = [1150*20(12)]*.851 = 1182.89
  20. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-20 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Trend Analysis Example with Seasonality:  Trend analysis was used to forecast the number of new hotel registrants (in ooo’s). The following data was used. yr1 yr2 1 Jan 17 17 2 Feb 16 15 3 Mar 16 17 4 Apr 25 24 5 May 24 23 6 June 25 25 7 July 23 24 8 Aug 20 19 9 Sep 20 20 10 Oct 16 15 11 Nov 16 15 12 Dec 17 17
  21. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-21 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 The trend analysis, using year1 data was Y= 20.5 + 0.1455X a) Compute the seasonal index b) Forecast July of year3, October of year3 c) What is the forecast for December if the average yearly demand for year is thought to increase by 10% higher than year1? Trend Analysis Example :
  22. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-22 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Using Regression Analysis to Forecast(Causal) Y Triple A' Sales ($100,000's) X Local Payroll ($100,000,000) 2.0 1 3.0 3 2.5 4 2.0 2 2.0 1 3.5 7
  23. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-23 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Using Regression Analysis to Forecast - continued Sales, Y Payroll, X X2 XY 2.0 1 1 2.0 3.0 3 9 9.0 2.5 4 16 10.0 2.0 2 4 4.0 2.0 1 1 2.0 3.5 7 49 24.5 S Y = 15.0 SX = 18 SX2 = 80 SXY = 51.5
  24. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-24 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Using Regression Analysis to Forecast - continued Calculating the required parameters: ( )( )( ) ( )( ) ( )( ) X . . Ŷ . . . X b Y a . . . X n X Y X n XY b . Y Y X X 25 0 75 1 75 1 3 25 0 5 2 25 0 6 80 5 2 3 6 5 51 5 2 6 15 6 3 6 18 6 3 2 2 2 + = = - = - = = - - = - - = = = = = = =    
  25. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-25 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Regression Equation 0 1 2 3 4 0 1 2 3 4 5 6 7 8 Area Payroll ($100,000,000) Triple A's Sale s ($100,000)
  26. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-26 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Methods to evaluate the Casual Regression Equation  Standard Error of the Estimate (the standard deviation)  Correlation Coefficient -1 < r <1  Coefficient of Determination 0 < r <1 the percent of variation in Y ( the dependent variable ) that is described by the X’s (independent variables ) 2
  27. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-27 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Standard Error of the Estimate - continued ( ) points data of number equation regression the from computed variable dependent the of value point data each of value = = - = - - =  n Y Y Y where n Y Y S c c X , Y 2 2 This is the standard deviation of the regression For Payroll example, S = 0.306 Y,X
  28. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-28 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Correlation Coefficient = ( ) [ ]   - 2 2 ( ) [ ]   - -  2 2 2 Y Y ( Y n X X n   -  Y X XY n r For Payroll example, r = 0.91
  29. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-29 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Coefficient - Four Values Fig. 5.7
  30. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-30 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Multiple Regression to Forecast #5-32 SUMMARY OUTPUT Regression Statistics Multiple R 0.656652082 R Square 0.431191956 Adjusted R Square 0.374311152 Standard Error 8.302983493 Observations 12 ANOVA df SS MS F Significance F Regression 1 522.6046512 522.6046512 7.580623398 0.020362831 Residual 10 689.3953488 68.93953488 Total 11 1212 Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Intercept 6.441860465 5.476681644 1.176234239 0.266742687 -5.760948796 18.64466973 # Tourists 1.23255814 0.447666868 2.753293191 0.020362831 0.235094026 2.230022253
  31. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-31 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Multiple Regression to Forecast #5-32 SUMMARY OUTPUT Regression Statistics Multiple R 0.673989793 R Square 0.454262242 Adjusted R Square 0.332987184 Standard Error 8.572787458 Observations 12 ANOVA df SS MS F Significance F Regression 2 550.5658367 275.2829184 3.745718626 0.065528166 Residual 9 661.4341633 73.49268481 Total 11 1212 Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Intercept 7.720611916 6.022703676 1.281917944 0.231907613 -5.903700728 21.34492456 Year -0.54576985 0.884817709 -0.616816147 0.552637609 -2.547368093 1.455828393 # Tourists 1.438808374 0.570482864 2.522088682 0.032656541 0.148285494 2.729331253
  32. To accompany Quantitative Analysis for Management, 7e by Render/ Stair 5-32 © 2000 by Prentice Hall, Inc. ,Upper Saddle River, N.J. 07458 Regression SAS printout Problem Attendance Wins 40,000 6 60,000 11 60,000 9 50,000 9 45,000 8 55,000 8 50,000 10 a) What is the dependent variable? b) Plot the data is it correlated?
Anúncio