Westminster International University in Tashkent

Semester 1, Academic Year 2023/2024

Quantitative methods
Seminar 8 Answers Time Series Forecasting
Task 1
b)

Chart Title
12,000

10,000

8,000 f(x) = 277.309523809524 x + 6662.60714285714

6,000

4,000

2,000

0
I II III IV I II III IV
1 2 3 4 5 6 7 8
2019 2020

a)

Number
Quarte of Seasonal
Year time r visitors Trend deviaton S=Y-T
2019 1 I 8 604 6939,91 1 664
2 II 6 556 7217,22 -661
3 III 3 824 7494,53 -3 671
4 IV 9 462 7771,84 1 690
2020 5 I 10 628 8049,15 2 579
6 II 8 275 8326,46 -51
7 III 4 881 8603,77 -3 723
8 IV 11 054 8881,08 2 173
c
2021 9 I . 9158,39
10 II 9435,7
11 III 9713,01
12 IV 9990,32

d)
Q1 Q2 Q3 Q4
2019 1 664 -661 -3 671 1 690
2020 2 579 -51 -3 723 2 173
Average deviation 2 121 -356 -3 697 1 932
Adjustment 0,005 0,005 0,005 0,005
Seasonal variation 2 121 -356 -3 697 1 932
Westminster International University in Tashkent
Semester 1, Academic Year 2023/2024

e) Y= Trend + Seasonal Variation = 9158,39 + 2121= 11 280

Task 2.
A, b, c is similar with the previous question.
d)
S=Y/T Q1 Q2 Q3 Q4
1,239786 2019 1,239786 0,908383 0,510239 1,217472
0,908383 2020 1,320388 0,99382 0,567309 1,244668
0,510239 Average deviation 1,280087 0,951101 0,538774 1,23107
1,217472 Adjustment 1,000258 1,000258 1,000258 1,000258
1,320388 Seasonal variation 1,279756 0,950856 0,538635 1,230753
0,99382
0,567309
1,244668
e) Y= Trend*Seasonal Variation = 9435.7* 1.28 = 11720,51. Forecasted number of visitors for 2021 Q1
using multiplicative model

Task 4. Multiplicative model

Chart Title
120

100

80

60

40

20

0
I II III IV I II III IV I II III IV I II III IV
2017 2018 2019 2020

Seasonal
Quarte Time trend deviation=Sales/Tre
Year r (x) Sales (y) nd
201
7 I 1 20 32,36 0,618041
36,04
II 2 30 6 0,832279
39,73
III 3 39 1 0,981604
IV 4 60 43,41 1,381973
Westminster International University in Tashkent
Semester 1, Academic Year 2023/2024

6
201 47,10
8 I 5 40 2 0,84923
50,78
II 6 51 7 1,004198
54,47
III 7 60 2 1,101481
58,15
IV 8 81 7 1,392772
201 61,84
9 I 9 50 3 0,808503
65,52
II 10 64 8 0,976682
69,21
III 11 74 3 1,069159
72,89
IV 12 95 9 1,30318
202 76,58
0 I 13 55 4 0,718167
80,26
II 14 68 9 0,847149
83,95
III 15 77 5 0,917163
IV 16 96 87,64 1,095393
202 68,79 91,32
1 I 17 8 5
87,50
18 5 95,01

Q1 Q2 Q3 Q4
2017 0,618041 0,832279 0,981604 1,381973
2018 0,84923 1,004198 1,101481 1,392772
2019 0,808503 0,976682 1,069159 1,30318
2020 0,718167 0,847149 0,917163 1,095393
average deviation 0,748485 0,915077 1,017352 1,293329
adjustment 0,993561 0,993561 0,993561 0,993561
seasonal variation 0,753336 0,921008 1,023945 1,301711

Question 5.
a) The number of patients at a dentist each day during a month will be time series data. True
b) A continuous increase of a number of subscribers to the internet provider “UzOnline” in
Uzbekistan is an example of trend component of a time series graph. False
c) The low sales of metro chips as a result of two hours of electricity loss are an example of
cyclical component of a time series graph for sales of metro chips during the day. False
d) The cyclical and irregular movements of a time series graph are always downward
movements of the time series graph. False
Westminster International University in Tashkent
Semester 1, Academic Year 2023/2024

e) The fundamental assumption of time-series analysis is that past patterns in time-series data
will continue unchanged in the future. True
f) The linear trend equation can be estimated by regression analysis. True
g) Time-series analysis generates forecasts by identifying cause and effect relationships
between variables. True