Time Series Analysis
Time Series Analysis
1
2
What is Time Series?
Any variable that is measured over a
successive point of time.
Numerical data obtained at regular time
intervals
Yearly, Monthly, Quarterly, Daily, and
Hourly.
3
Time Series example
Yearly numbers of enrollees
Monthly number of Population
Quarterly sales of a company product
4
What is the use of Time Series Analysis?
16.00
14.00
12.00
10.00
8.00
6.00
4.00
2.00
0.00
1975
1977
1979
1981
1983
1991
1993
1995
1997
1999
2001
1985
1987
1989
Y e ar
6
Time-Series Components
Time-Series
7
Trend Component
Overall or persistent, long-term upward
or downward pattern of movement.
Changes in Technology, Consumers
reference, Population.
Observe in several years.
8
Trend Component
rd t rend
Sales U pw a
Time
9
Trend Component Pattern(continued)
Trend can be upward or downward linear
Sales Sales
Time Time
Downward linear trend Upward linear trend
10
Trend Component Pattern
Trend can be linear or non-linear
Sales
Sales
Time Time
Downward nonlinear trend Upward nonlinear trend
11
Trend Component Pattern
No trend
It has no consistent increase and decrease
Sales
Time
No trend
12
Seasonal Component
Fairly regular periodic fluctuations.
Weather conditions
Social customs
Observe within 12 months (Quarterly or
Monthly)
Seasonal Component
Repetition
of Pattern
14
Cyclical Component
Repeated up and down swings
Regularly occur but may vary in length
Often measured by Peak (Prosperity) and
Contraction (Recession).
Results of a long term factors influencing
economy.
Cyclical Component
Sales
Peak Peak
Contraction
Years
Irregular Component
Unpredictable, random, “residual”
fluctuations
Due to the following reasons:
Strikes
Hurricanes or any natural occurrences
“Noise” in the time series
Irregular Component
Sales
Unpredictable
Time
18
Smoothing Techniques
To produce a better forecast we need to
determine which components are present in a
time series.
To identify the components present in the
time series, we need first to remove the
random variation.
This can be done by smoothing techniques.
19
Moving Averages
• The Term Moving indicates that as new
observation is available, it replace the old
observation.
• L or the Length to identify the integer value of the
average length of the data.
20
Moving Average: Example
Week Sales (1000s of
12 Weeks Gallons)
Gasoline Sales 1 17
2 21
3 19
4 23
5 18
6 16
7 20
8 18
9 22
10 20
11 15
12 22
21
Moving Average: Gasoline Sales Graph
22
Moving Average: Example
Week Sales (1000s of
Gallons)
MA(L) = S1 + S2 + S3
L 1 17
2 21
3 19
4 23
MA(3) = 17 + 21 + 19 = 19 5 18
6 16
3 7 20
8 18
9 22
10 20
11 15
12 22
23
Moving Average: 3 period
Week Sales (1000s of
Gallons)
1 17
2 21 3 period moving
3 19 average
4 23 19 (17+21+19)/3 = 19
5 18 21 (21+19+23)/3 = 21
6 16 20
7 20 19
8 18 18
9 22 18
10 20 20
11 15 20 Predicted Value
12 22 19
13 19
24
Moving Average: 3 period moving period
moving average graph
25