FORECASTING

You might also like

Download as pdf
Download as pdf
You are on page 1of 24
CHAPTER 9 Forecasting cast is a prediction, wh , . estima Hain set of le, or di pased ona cer of factor leterminati i? fynds. GNP. Technological statue The value being lore Of what wil occur in the oe oye ay ot te folowing: Ws nd s0 on. The forges may be soles, stsones Siyea paren teoted tote. er Brn or door) cerca ¢ | company data, or _categoies of Forecasting Time Hoi ns term Forecast. It covers or ne doy Y Yo one year and are used mainly for short id areiusad short-run such as employment, purchasing, scheduling, sales, and production rates. 1g, scheduli le iu watrol i diate-term Forecast. t. A period ranging from one season fo one or f two years », Interme' » oad is used for production schedul les, revenues, cash flow, 7 and budget plannin. 9 /1ong-term Forecast. i 3/Long- A period which spans from two to five years or more I | facilities expansion, and general policy. cre ee mas 3 fends. technology syecasting Techniques qualitative Techniques: © technique that is based ‘on qualitative dato he aggregate opinion of tne sales force to forecast the future a 1k statistical forecasting techriaue that is bosed solely 07 rime Series Analysis. ed over a period of time. 2 historical data accumulat 3, Causal Methods. * method that define °° ielationsrips among independent ond dependent variables ‘system of elated equations. forecasting Methods O° rime series Methods 1, simple Moving average 2. Weighted Moving ‘Average 3. simple Exponentic moothing 4, Adjusted ‘sential smoo" 9 5. Forecast Reliability - ene 08 wen need pile eI Fon cor dng the nex! Bamps Bete oe cial ea sock for recordet tne ee forecast demand for the Be cto aco in kored I distributor is i cars is not or” jer following data for Fas Month Januar Februar March April May June Jul August September g@aogo z a = a g 3 Solution: 9.1 Simple Moving Average © Simple Moving Average is the see yEiGhIed average of a consecy data points, It can be used as a forecast seasonal ‘Adjustment of the data, t Simple Moving Average = Zio recent n dota volvey ive numberof 9.1.1 Computation of 3-Month Moving Average April Moving Average = S++ 90 « ‘0 = 60 May Moving Average - Rss 90 June Moving Average = 0+ 9 +10 July Moving Average = Reto + 9 August Moving Average = e894 180 September Moving Average = 82+ 150 +70 moving Average = 150+ 70 2 )+ 110 -80 S110 2 ot moving Average = 041104150 oy 3 “7 wo , Moving Average = 10 +150 +130 eo Bi | gl Computation of 5-Month Moving Average moving Average = 60+70+50+90+10 _ 260 3 56 wy wowing Average = 70 +50+90+10+80 300 _ yo 5 == 5 gst Moving Average = 5049010 04150. on on seplerber Moving Average = rosin | * 0 a he 5 ober Moving Average = 102206 1D.705100 420 _ a4 m od woyernber Moving Average = sorisiemsri@ Ee 1-701 8 pecember Moving Average = Three- and Five-Month Moving Average Table 9.1: Failed] Month | Motorcycles $-Month Moving | 5-Month Moving Demanded | Average Average 1 January 60 2 February 70 3 March 50 z 4 April 90 3 May 10 70. =z 6 une | 50 7] 150, [of 7 [ly “7 80 76 8 8 ip 80 [80 9 | September [0 +—jig__| 0 150 | 2 — 84 19_| Octobe 30 no n12 Novembet [190 +79 122 tw 12_| December 2 Figure 9.1: Three-and Five-Month Moving Averages 9.2 Weighted Moving Average < Itisa time series forecasting method in which th Smoothing Constant is a weighting factor used in the technique. © most recent data are weighteg exponential smoothing forecasting The formula for the weighted moving average is: Wr = Wx, where: Wh = weight moving average W. => weight for period i Xi = data in period i Computation of Weighted Moving Average (WMA) April WMA = 20%(60) + 30%(70) + 50%(50) =12+21+25= 5g May WMA = 20%(70) + 30%(50) + 50%(90) =14+15+45= 74 June WMA = 20%(50) + 30%(90) +50%(10) =104+27+ 5= 42 July WMA = 20%(90) + 30%(10) + 50%(80) = 18+ 3+40= 61 August WMA = 20%(10) + 30%(80) + 50%(150) = 2424+75= 101 September WMA = 20%(80) + 30%(150) + 50%|70) =16+45+35= 96 October WMA 20%(150) + 30%(70) + 50%(110) = 30 +21 +55 = 106 November WMA 0% (70) + 30%(110) + 50%(150) = 14 +33 + 75 = 129 December WMA —— = 20%(110) + 30%(150) + 50%(130) = 22 + 45 + 65 = 132 272 September 2 i October ie po November | —i3p = December A simple exponential Smoothing s His weighted moving average technique aio. The “ange of smoothing 3. I belween D0 ond ba iis ig denied the formula for the weighted moving average is: fe at (1-0) Fi = forecost for the next petiod > gctval data in he presen! peed 4, he previousy determined eens fore preset pod weighting factrteered 10 the smootingcorsont where: Fret Xi fot a>a r exponent fe a = 0.10, therefore (4-0) = 090. Pen subse fo the exPore twe will Us smoothing formula. 2 ous neaYt (1-0) Fe reall sian = (0.10)(70) + (0.901160) 20 =7 +54 =e =61 Ys accel heaYet(1-a) Fa fre 0127 = (0.10)(90) + (0.901599) =94+53.91 = 62.91 FreaYet(1-a)Fo = (0.10}(80) + (0.90)(57.62) =8+ 51.86 = 59.86 Fo=aYet(1-a)Fs = (0.10)(70) + (0.90) (68.87) =7 +6198 = 68.98 Fi =a Yo + (1-4 Fro = (0.10}(150) + (0.90)(73.08) =15+65.77 = 80.77 Fee aYr+(1-a)Fr (0.10)(150) + (0.90)(59.86) = 15+ 53.87 = 68.87 Fio= aYot(1-a) Fo (0.10)(110) + (0.90)(68.98) 11 + 62.08 73.08 aYy+(1-a) Fis solution of exponential smoothing when a = 0.30, therefore (1~a)= 0.70, then substitu to simple exponential smoothing Frei = a Yi (1a) Fr Fs=aYo+(1-a)Fo = (0.30}(70) + (0.70) (60) = 21+ 42 =63 Fs=aYet(1-0) Fy = (0.30)(90) + (0.70)(59.1) =27 +4137 = 6837 FreaYet(1-a)F = (0.30)(80) + (0.70)(50.86) = 24+ 35.60 | = 59.40 Fo=aYe+(1-a)Fy = (0.30) (70) + (0.70)(86.72) =21 +6070 = 68.98 Fu=aYot(1-a Fo | = (0.30)(150) + (0.70}(90.19) = 45 + 63.13 = 108.13 Fe= aYs+(1-a) Fo = (0.30)(50) + (0.70}(63) = 154441 = 59.1 Fe= aYs+(1-a)Fs = (0.30}(10) + (0.70)(68.37) = 3+ 4786 = 50.86 fa= aYr+(1~-a) Fr = (0.30)(150) + (0.70) (59.60) = 4544172 = 86.72 Fio= aYg+(1—a) Fo = (0.30)(110) + (0.70)(81.70) = 33457.19 = 90.19 Fas ay +( 1-0) Fis = (0.30)(130) + (0.70)(108.13) . = 39+ 75.69 = 114.69 214 table 9.3 Exponential smoothing Forecast month | _ Motorcycles] | Feseatis— Demanded yy | —-—FoHeeasl Fa tty Fae | x —_ 3} S| - 63 ff 10 oN a £0, 57.62 Oe | 130 39.86 > 2 | 7” £887 8672 WW £8.58 81.70 | 150. 73.08 019 130 80.7 108.13 i 35.67 | 1469 | } Jing Forecast . Exponential Smoothi lied 9 ah Pe 0.10 and a = 0.30 Jnanges ond {4 Adjusted Forecast — sie wrondie it stin i thing forec The exponential smo ee i dusted forecast 0.80 substitute to a )=08 hha Adjusted Raa fa tl Ling B = 0.20, therefore (1-8 Ta = Big f+ 0-R Tye = (0.20) Fev — Fy) + 0.801 275 ) I= pl Fa =F) + (1-BIT ro 1-6 adjusted Fn = Fi +5 where: AFisi => Adjusted expr T = an exponenti 1- 0.20 Afi = Far CaP}. “Fasat, onential smoothing in the present period ially smoothed trend factor r= the last period trend factor Let T2=0 0.20 ( Fs- Fa) + 0.80T2 0.20 ( 61 - 60) +0.80(0) =0.20(1)+0 =0.20+0 =0.20 Ts = 0.20 | Fa- Fs) + 0.8013 = 0.20 (59.9 - 61) +0.80 (0.20) =0.20(-I.1) 0.16 =-0.22+0.16 = 0.06 Ts=0.20( Fs~Fa) + 0.80T« 20 ( 62.91 - 59.9) + 0.80 (-0.06 ) 0.20 ( 3.01 ) - 0.048 0.602 - 0.048 =0.55 Te= 0.20 | Fe—Fs) + 0.80Ts = 0.20 ( 57.62- 62.91 } +0.80 (0.55) 0.20 (-5.29) + 0.44 058 + 0.44 62 17 = 0.20 | Fr—Fs) + 0.801. 20 | 59.86~ 57.62) + 0.80 | -0.42) .20 2.24 ) -0.496 .20 ( Fs~ Fr) + 0.801) 20 | 68.87 ~ 59.86 } +04 20 (9.01) -0.04 }+ 0.80 (-0.05 ) 802-0.04 76 AFS= Fs Ts =61+4(0.20) AFa AFs AFs = 59.86 + 4(-0.05) = 59.86 -0.20 = 59.66 AFe = Fa + 47g = 68.87 +4( 1,76) = 68.87 +7.04 = 7591 276 ono Fr 1 Oa 0150 ( 68:98 - 68.87 ) +0 950 (0.11) + 1.408 80 (176) £9022 + 1.408 2143 4¢20:20 (Fro~ Fr) + 0.801» = 0:90 ( 73:08 ~ 68.98) + 0.80 0.20 (4.1) + 1-144 (143) 82+ 1.144 96 20.20 ( Fir~ Fro) + 0.800 2.0.20 ( 80.77 - 73.08 } + 0.80 (1.96) 0.20 ( 7.69) + 1.568 82 + 1.568 3.11 Tio 0.20 ( Fr2- Fn) + 0.8070 0.20 ( 85.69 - 80.77 ) + 0.80( 3.11] = 0.20 ( 4.92) + 2.488 20.984 + 2.488 =3.47 Table 9.4: Adjusted Expone! Afr = Fr + Aly = 6B9B+ 41 143 = 68.98 +572 " =747 AFio= Fio + 47 3.08 + 4( 1:96) 3.08 + 7A = 80.92 AB) = Fin #41 80.7744 (3.11) 80,77 + 12.44 = 93.21 AFia= Fiat Alia = 85.69 + 4(3.47) = 80.77 + 13.88 299.57 ntial Smoothing Forecost ‘Motorcycles Forecast, Demanded (Y') Fret Forecee B=0. ‘Adjusted | | nf —a | 59.66 51

You might also like