CH 9
CH 9
CH 9
DEVELOPED BY
ORE A. SOLUADE
Chapter 9: FORECASTING
LIST OF WORKSHEETS
TOC Table of Contents
3-per.MA Table of 3-period moving average calculations
3-per.MA_Chart Graph of 3-period moving average calculations
3-per.MA_TS(t) Tracking Signal for 3-period moving average calculations
3-per.WMA Table of 3-period Weighted Moving Average calculations
3-per.WMA_Chart Graph of 3-period Weighted Moving Average calculations
3-per.WMA_TS(t) Tracking Signal for 3-period Weighted Moving Average calculations
4-per.WMA Table of 4-period Weighted Moving Average calculations
4-per.AMA_Chart Graph of 4-period Weighted Moving Average calculations
4-per.WMA_TS(t) Tracking Signal for 4-period Weighted Moving Average calculations
5-per.WMA Table of 5-period Weighted Moving Average calculations
5-per.AMA_Chart Graph of 5-period Weighted Moving Average calculations
5-per.WMA_TS(t) Tracking Signal for 5-period Weighted Moving Average calculations
Exp.Smthng Table of Exponential Smoothing calculations
Exp_Chart Graph of exponential Smoothing calculations
Exp_TS(t) Tracking Signal for Exponential Smoothing calculations
Regression Table of Regression calculations
Reg_Chart Graph of Linear Regression calculations
Page 1
3-per.MA
OBJECTIVE:
This template is used to make forecasts based on a 3-period Moving Average technique.
The 3-period moving average calculation is for a time series of 30 periods.
By entering data in column B, error measurements, Tracking Signal, as well as the
charts, are automatically generated.
INPUT:
(At - Ft)2 |At - Ft|
At Ft At - Ft Error Absolute
Period Actual Forecast Error Squared Deviation RSFE MAD(t) TS(t)
1 800
2 1400
3 1000
4 1500 1067 433 187778 433 433 433 1
5 1500 1300 200 40000 200 633 317 2
6 1300 1333 -33 1111 33 600 222 3
7 1800 1433 367 134444 367 967 258 4
8 1700 1533 167 27778 167 1133 240 5
9 1300 1600 -300 90000 300 833 250 3
10 1700 1600 100 10000 100 933 229 4
11 1700 1567 133 17778 133 1067 217 5
12 1500 1567 -67 4444 67 1000 200 5
13 2300 1633 667 444444 667 1667 247 7
14 2300 1833 467 217778 467 2133 267 8
15 2000 2033 -33 1111 33 2100 247 8
16 1700 2200 -500 250000 500 1600 267 6
17 1800 2000 -200 40000 200 1400 262 5
18 2200 1833 367 134444 367 1767 269 7
19 1900 1900 0 0 0 1767 252 7
20 2400 1967 433 187778 433 2200 263 8
21 2400 2167 233 54444 233 2433 261 9
22 2600 2233 367 134444 367 2800 267 11
23 2000 2467 -467 217778 467 2333 277 8
24 2500 2333 167 27778 167 2500 271 9
25 2600 2367 233 54444 233 2733 270 10
26 2200 2367 -167 27778 167 2567 265 10
27 2200 2433 -233 54444 233 2333 264 9
28 2500 2333 167 27778 167 2500 260 10
29 2400 2300 100 10000 100 2600 254 10
30 2100 2367 -267 71111 267 2333 254 9
Page 2
3-per.MA
OUTPUT:
a Bias 86.41975
b Mean Square Error 91440.33
c Standard Error 302.391
d Mean Absolute Deviation 254.321
e Tracking signal See Column I
f Graph Click on 3-per.MA_Chart Tab
Page 3
3-per.MA_Chart
3000
2500
2000
Demand
1500
1000
500
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
Period
Actual Forecast
Page 4
500
3-per.MA_Chart
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
Period
Actual Forecast
Page 5
3-per.MA_TS(t)
12
10
8
Tracking Signal
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Period
Page 6
2
3-per.MA_TS(t)
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Period
Page 7
3-per.WMA
OBJECTIVE:
This template is used to make forecasts based on a 3-period Weighted Moving Average technique.
The 3-period Weighted Moving Average calculation is for a time series of 30 periods.
The weights: a1, a2, and a3 are entered as required.
By entering data in column B, error measurements, Tracking Signal, as well as the
charts, are automatically generated.
INPUT:
a1 = 0.40
a2 = 0.35
a3 = 0.25
(At - Ft)2 |At - Ft|
At Ft At - Ft Error Absolute
Period Actual Forecast Error Squared Deviation RSFE MAD(t) TS(t)
1 800
2 1400
3 1000
4 1500 1060 440 193600 440 440 440 1
5 1500 1285 215 46225 215 655 328 2
6 1300 1300 0 0 0 655 218 3
7 1800 1450 350 122500 350 1005 251 4
8 1700 1505 195 38025 195 1200 240 5
9 1300 1575 -275 75625 275 925 246 4
10 1700 1640 60 3600 60 985 219 4
11 1700 1560 140 19600 140 1125 209 5
12 1500 1540 -40 1600 40 1085 191 6
13 2300 1650 650 422500 650 1735 237 7
14 2300 1780 520 270400 520 2255 262 9
15 2000 1980 20 400 20 2275 242 9
16 1700 2225 -525 275625 525 1750 264 7
17 1800 2045 -245 60025 245 1505 263 6
18 2200 1845 355 126025 355 1860 269 7
19 1900 1860 40 1600 40 1900 254 7
20 2400 1965 435 189225 435 2335 265 9
21 2400 2145 255 65025 255 2590 264 10
22 2600 2200 400 160000 400 2990 272 11
23 2000 2450 -450 202500 450 2540 281 9
24 2500 2370 130 16900 130 2670 273 10
25 2600 2365 235 55225 235 2905 272 11
26 2200 2325 -125 15625 125 2780 265 10
Page 8
3-per.WMA
OUTPUT:
a Bias 93.7037
b Mean Square Error 93692.59
c Standard Error 306.0925
d Mean Absolute Deviation 254.8148
e Tracking signal See Column I
f Graph Click on 3-per.WMA_Chart Tab
Page 9
3-per.WMA_Chart
3000
2500
2000
Demand
1500
1000
500
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Period
Actual Forecast
Page 10
500
3-per.WMA_Chart
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Period
Actual Forecast
Page 11
3-per.WMA_TS(t)
12
10
8
Tracking Signal
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
Period
Page 12
2
3-per.WMA_TS(t)
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
Period
Page 13
4-per.WMA
OBJECTIVE:
This template is used to make forecasts based on a 4-period Weighted Moving Average technique.
The 4-period Weighted Moving Average calculation is for a time series of 30 periods.
The weights: b1, b2, b3 and b4 are entered as required.
By entering data in column B, error measurements, Tracking Signal, as well as the
charts, are automatically generated.
INPUT:
b1 = 0.40
b2 = 0.30
b3 = 0.20
b4 = 0.10
(At - Ft)2 |At - Ft|
At Ft At - Ft Error Absolute
Period Actual Forecast Error Squared Deviation RSFE MAD(t) TS(t)
1 800
2 1400
3 1000
4 1500
5 1500 1090 410 168100 410 410 410 1
6 1300 1310 -10 100 10 400 210 2
7 1800 1280 520 270400 520 920 313 3
8 1700 1490 210 44100 210 1130 288 4
9 1300 1520 -220 48400 220 910 274 3
10 1700 1530 170 28900 170 1080 257 4
11 1700 1660 40 1600 40 1120 226 5
12 1500 1580 -80 6400 80 1040 208 5
13 2300 1520 780 608400 780 1820 271 7
14 2300 1720 580 336400 580 2400 302 8
15 2000 1820 180 32400 180 2580 291 9
16 1700 1950 -250 62500 250 2330 288 8
17 1800 2180 -380 144400 380 1950 295 7
18 2200 2040 160 25600 160 2110 285 7
19 1900 1890 10 100 10 2120 267 8
20 2400 1850 550 302500 550 2670 284 9
21 2400 2000 400 160000 400 3070 291 11
22 2600 2170 430 184900 430 3500 299 12
23 2000 2220 -220 48400 220 3280 295 11
24 2500 2400 100 10000 100 3380 285 12
Page 14
4-per.WMA
OUTPUT:
a Bias 122.6923
b Mean Square Error 100457.7
c Standard Error 316.9506
d Mean Absolute Deviation 248.0769
e Tracking signal See Column I
f Graph Click on 4-per.WMA_Chart Tab
Page 15
4-per.WMA_Chart
3000
2500
2000
Actual
Sales
1500 Forecast
1000
500
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
PeriodPage 16
500
4-per.WMA_Chart
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Period
Page 17
4-per.WMA_TS(t)
16
14
12
10
Tracking Signal
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
Period
Page 18
2
4-per.WMA_TS(t)
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
Period
Page 19
5-per.WMA
OBJECTIVE:
This template is used to make forecasts based on a 5-period Weighted Moving Average technique.
The 5-period Weighted Moving Average calculation is for a time series of 30 periods.
The weights: c1, c2, c3 c4 and c5 are entered as required.
By entering data in column B, error measurements, Tracking Signal, as well as the
charts, are automatically generated.
INPUT:
c1 = 0.30
c2 = 0.25
c3 = 0.20
c4 = 0.15
c5 = 0.10
(At - Ft)2 |At - Ft|
At Ft At - Ft Error Absolute
Period Actual Forecast Error Squared Deviation RSFE MAD(t) TS(t)
1 800
2 1400
3 1000
4 1500
5 1500
6 1300 1165 135 18225 135 135 135 1
7 1800 1325 475 225625 475 610 305 2
8 1700 1350 350 122500 350 960 320 3
9 1300 1525 -225 50625 225 735 296 2
10 1700 1520 180 32400 180 915 273 3
11 1700 1545 155 24025 155 1070 253 4
12 1500 1650 -150 22500 150 920 239 4
13 2300 1580 720 518400 720 1640 299 5
14 2300 1610 690 476100 690 2330 342 7
15 2000 1810 190 36100 190 2520 327 8
16 1700 1890 -190 36100 190 2330 315 7
17 1800 1955 -155 24025 155 2175 301 7
18 2200 2100 100 10000 100 2275 286 8
19 1900 2020 -120 14400 120 2155 274 8
20 2400 1905 495 245025 495 2650 289 9
21 2400 1925 475 225625 475 3125 300 10
22 2600 2070 530 280900 530 3655 314 12
23 2000 2235 -235 55225 235 3420 309 11
Page 20
5-per.WMA
OUTPUT:
a Bias 135.8
b Mean Square Error 109251
c Standard Error 330.5314
d Mean Absolute Deviation 277.8
e Tracking signal See Column I
f Graph Click on 5-per.WMA_Chart Tab
Page 21
5-per.WMA_Chart
3000
2500
2000
Actual
Sales
1500 Forecast
1000
500
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
PeriodPage 22
500
5-per.WMA_Chart
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Period
Page 23
5-per.WMA_TS(t)
14
12
10
8
Tracking Signal
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Period
Page 24
2
5-per.WMA_TS(t)
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Period
Page 25
Exp.Smthng
OBJECTIVE:
This template is used to make forecasts based on the Exponential smoothing technique
for a time series of 30 periods. The value of the smoothing constant in entered in cell B15.
By entering data in column B, error measurements, Tracking Signal, as well as the
charts, are automatically generated.
INPUT:
alpha = 0.3
(At - Ft)2 |At - Ft|
At Ft At - Ft Error Absolute
Period Actual Forecast Error Squared Deviation RSFE MAD(t) TS(t)
1 800 800 N/A N/A N/A N/A N/A N/A
2 1400 800 600 360000 600 600 600 1
3 1000 980 20 400 20 620 310 2
4 1500 986 514 264196 514 1134 378 3
5 1500 1140 360 129456 360 1494 373 4
6 1300 1248 52 2689 52 1546 309 5
7 1800 1264 536 287620 536 2082 347 6
8 1700 1425 275 75851 275 2357 337 7
9 1300 1507 -207 42937 207 2150 321 7
10 1700 1445 255 65000 255 2405 313 8
11 1700 1522 178 31850 178 2584 300 9
12 1500 1575 -75 5636 75 2509 279 9
13 2300 1553 747 558679 747 3256 318 10
14 2300 1777 523 273753 523 3779 334 11
15 2000 1934 66 4389 66 3845 315 12
16 1700 1954 -254 64326 254 3592 311 12
17 1800 1878 -78 6012 78 3514 296 12
18 2200 1854 346 119525 346 3860 299 13
19 1900 1958 -58 3363 58 3802 286 13
20 2400 1941 459 211053 459 4261 295 14
21 2400 2078 322 103416 322 4583 296 15
22 2600 2175 425 180717 425 5008 302 17
23 2000 2302 -302 91460 302 4706 302 16
24 2500 2212 288 83119 288 4994 302 17
25 2600 2298 302 91091 302 5296 302 18
26 2200 2389 -189 35620 189 5107 297 17
27 2200 2332 -132 17454 132 4975 291 17
28 2500 2292 208 43065 208 5182 288 18
29 2400 2355 45 2049 45 5228 279 19
Page 26
Exp.Smthng
OUTPUT:
a Bias 171.0138
b Mean Square Error 111266.1
c Standard Error 333.5658
d Mean Absolute Deviation 278.8086
e Tracking signal See Column I
f Graph Click on Exp_Chart Tab
Page 27
Exp_Chart
Exponential Smoothing
3000
2500
2000
Demand
1500
1000
500
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Period
Actual Forecast
Page 28
500
Exp_Chart
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Period
Actual Forecast
Page 29
Exp_TS(t)
Exponential Smoothing
20
18
16
14
12
Tracking Signal
10
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
Period
Page 30
4
2 Exp_TS(t)
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
Period
Page 31
Regression
OBJECTIVE:
This template is used to make forecasts based on Regression Analysis for a
30-period data set.
By entering data in column B, error measurements, Tracking Signal, as well as the
charts, are automatically generated.
INPUT:
(At - Ft)2 |At - Ft|
Actual Trend At Ft At - Ft Error Absolute
t At t*At t2 A t
2
Tt t Actual Forecast Error Squared Deviation RSFE
1 600 600 1 360000 801.3 1 600 801.3 -201.3 40514.60 201.28 -201.28
2 1550 3100 4 2402500 1160.9 2 1550 1160.9 389.1 151400.52 389.10 187.82
3 1500 4500 9 2250000 1520.5 3 1500 1520.5 -20.5 420.79 20.51 167.31
4 1500 6000 16 2250000 1880.1 4 1500 1880.1 -380.1 144497.75 380.13 -212.82
5 2400 12000 25 5760000 2239.7 5 2400 2239.7 160.3 25681.99 160.26 -52.57
6 3100 18600 36 9610000 2599.4 6 3100 2599.4 500.6 250641.01 500.64 448.07
7 2600 18200 49 6760000 2959.0 7 2600 2959.0 -359.0 128862.91 358.97 89.10
8 2900 23200 64 8410000 3318.6 8 2900 3318.6 -418.6 175217.76 418.59 -329.49
9 3800 34200 81 14440000 3678.2 9 3800 3678.2 121.8 14833.88 121.79 -207.70
10 4500 45000 100 20250000 4037.8 10 4500 4037.8 462.2 213609.43 462.18 254.48
11 4000 44000 121 16000000 4397.4 11 4000 4397.4 -397.4 157955.69 397.44 -142.95
12 4900 58800 144 24010000 4757.1 12 4900 4757.1 142.9 20434.19 142.95 -0.01
78 33350 268200 650 112502500 28491.7 0.0 1324070.51 3553.85
t bar = 6.5
A bar = 2779.2
b(1) = 359.62
b(0) = 441.67
Tt = b(0) + b(1)*t
= 441.67+ 359.62*t
OUTPUT:
a Bias -0.00043
b Mean Square Error 132407
c Standard Error 363.878
d Mean Absolute Deviation 355.385
e Tracking signal See Column P.
f Graph Click on Reg_Chart Tab
Page 32
Regression
MAD(t) TS(t)
201.28 -1.00
295.19 0.64
203.63 0.82
247.76 -0.86
230.26 -0.23
275.32 1.63
287.27 0.31
303.69 -1.08
283.48 -0.73
301.35 0.84
310.08 -0.46
296.15 0.00
Page 33
Reg_Chart
Regression Analysis
6000
5000
4000
Sales
3000
2000
1000
0
1 2 3 4 5 6 7 8 9 10 11 12
Period
Actual Trend
Page 34
1000
Reg_Chart
0
1 2 3 4 5 6 7 8 9 10 11 12
Period
Actual Trend
Page 35