Time series analysis is not an easy topic, I must say. At least for me. During my graduation, I had the chance to learn the subject, and it took me some time to get my head around it. The purpose of this article is to energize what I have learned and trying to explain to others.
You will enjoy time series analysis if you know the real-life use of time series. Let’s see the use of time series in real life.
Time Series in Real Life:
- It enables us to study the past behaviour of the phenomenon under consideration, i.e., to determine the type and nature of the variation in the data
- There are apparent forecasting purposes in sales and demand, but another useful area is in statistical process control for the chemical and process manufacturers
- It enables us to predict or estimate the behaviour of the phenomenon in the future which is essential for business planning
However, other aspects come into play when we are dealing with time series. So we will enter the components of the time series.
Components of the Time Series:
Time series has four elements that used to identify the patterns of the data. But it is not compulsory to present trends so let us discuss each of them in detail now.
The word trend means ‘tendency’. The general tendency of the time series data is to increase or decrease during a long period is called the secular trend or simple trend. This phenomenon observed in time series relating to Economics and Business, For example, an upward tendency found in time series relating to population, production, and sales of the product, etc., while a downward trend noticed in the time series relating to deaths, epidemics, etc.
“The trend also called a secular or long-term trend is the basic tendency of a series to grow or decline over some time. The concept of trend does not include short-range oscillations, but rather the steady movement over a long time.”
Linear and Non-Linear Trend: If the time series data plotted on the graph more or less round a straight line, the trend exhibited by time series termed as Linear otherwise Non-Linear
In a straight-line trend, the time-series data increase or decrease more or less by a constant amount.
A seasonal pattern exists when a series is affected by seasonal factors, i.e., during 12 months and have the same or almost the same pattern year after year. Thus, seasonal variation in a time series will be there if the data recorded quarterly(every three months), monthly, weekly, daily, hourly, and so on.
Seasonal fluctuations influence most of the economic time series, e.g., prices, production, sales and profits in a departmental store, etc., are all affected by seasonal variations.
The seasonal variations may be associated with the following two causes:
- Resulting from natural forces: As the name hints, the various seasons or weather conditions and climate changes play an important role in seasonal changes. For example, the sales of the umbrella in the rainy season, the demand for electric coolers in summer, etc.
- Resulting from human-made conventions: These variations in a time series within 12 months are due to habits, fashions, customs and conventions of the society. For example, sales of jewellery and ornaments go up considerably during marriages.
The oscillatory movements in a time series with a period of oscillations greater than one year termed as cyclical variations. These variations in a time series are due to ups and downs recurring after a period higher than one year.
It is a combination of cyclical and seasonal variations, it is essential in every time series another factor called random variations. Such variations do not display any definite pattern, and there is no regular period or time of their occurrence. Hence they named irregular variations.
This was just an introduction to Time Series. We will see the details in the part-2. See you next time.