Weekday
Mon
Tue
Wed
Thu
Fri
Mon
Tue
Wed
Thu
Fri
Demand
75
80
99
108
65
80
78
104
115
68
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
1. Create a forecasting model
for this data, using the
Winter's Seasonal Model.
2. Update this model with
new information which will be
given to you after you finish
the first model.
140
120
100
80
60
40
20
0
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
STEP 1
STEP 2A
Observations
t
Day
-9 Monday
-8 Tuesday
-7 Wednesday
-6 Thursday
-5 Friday
-4 Monday
-3 Tuesday
-2 Wednesday
-1 Thursday
0 Friday
x
Centered
Moving
Average
STEP 3
STEP2B
First estimate
Normalized
of Seasonal Seasonal Factor Deasonalized
Factor
Estimate
Data
Average of
Seasonal Factor
Estimates
Normalized
Seasonal Factor
Estimate
t
75
80
99
108
65
80
78
104
115
68
85.4
86.4
86
87
88.4
89
1.159
1.250
0.756
0.920
0.882
1.169
0.925
0.887
1.171
1.257
0.760
0.925
0.887
1.171
1.257
0.760
81.10
90.15
84.58
85.91
85.51
86.51
87.90
88.85
91.48
89.46
Monday
Tuesday
Wednesday
Thursday
Friday
0.920
0.882
1.164
1.250
0.756
4.972
0.925
0.887
1.171
1.257
0.760
5.000
Weekday t
Mon
-9
Tue
-8
Wed
-7
Thu
-6
Fri
-5
Mon
-4
Tue
-3
Wed
-2
Thu
-1
Fri
0
-45
𝑥𝑡′
81.10
90.15
84.58
85.91
85.51
86.51
87.90
88.85
91.48
89.46
871.44
t*x
-729.893
-721.214
-592.033
-515.456
-427.558
-346.024
-263.694
-177.695
-91.4775
0
-3865.04
t2
92.00
81
64
49
36
25
16
9
4
1
0
285
y = 0.6838x + 90.221
90.00
88.00
86.00
84.00
82.00
80.00
-10
Use linear regression
b0
a0
-8
-6
-4
-2
0
0.6838
90.22
Forecasts: Monday:
Tuesday:
84.07
81.27
Smoothing Constants:
a
0.2
b
g
0.1
0.3
Monday:
89 <-- This will be given to you after completion of initialization
Update:
𝑎
ෞ1
෢
𝑏1
𝐹෡1 ′
91.971 < Update the model parameters.
0.7905
0.9377
𝐹෡𝑡 ′
𝐹෡𝑡
0.938
0.887
1.171
1.257
0.760
0.935 <-- Normalize all the factors
0.885
1.168
1.254
0.758
5.013
5.000
Forecast Tuesday:
Wednesday:
82.1
109.2
Mon
Tue
Wed
Thu
Fri