Major Components of Time Series Analysis

There are four different components of a time series, namely:

Components of Time Series

1. Secular Trend
2. Seasonal Variation
3. Cyclical Variation
4. Irregular Variation

Secular Trend

It measures long term changes occurring in a time series without bothering about any short-term variations occurring in between. In other words, secular trend measures the smooth and regular long-term movements.

The graphs show trend like straight line running from left bottom to right top, left top to right bottom or parallel to abscissa depicting growth or decline or stagnation respectively. In some cases, curvy linear trend also studied.

Seasonal Variation

Seasonal variation measures the short-term fluctuations observed in a time series data, specifically in a particular year. For example, particular item or product has more sales in a particular season, like rain coats in rainy season etc.…

All these types of variations are coming under seasonal variation. Season variations are more akin to climatic or weather conditions.

Cyclic Variation

Cyclic variation relates to periodic changes, particularly in a business. A cycle contains more than a year period. For an example cycle variation is like ups and downs in business, recession, recovery, etc.

Large number of factors responsible for occurrence of cycles,

1. Certain periods likes and dislikes
2. Production of certain product stopped and new product launched. Again, old product are adopted. Such changes form cycles.
3. New scientific and technology developments

Normality Test in R

Irregular Variations

Irregular variation also called as random variations. It is not possible to think of them of occurrence, directions and magnitude. The variation is occurring basis of earthquakes, floods, wars etc.

Irregular variation also known as in another name is residual variation.

In classical time series analysis, mathematical model is the multiplicative model.

Usually, Y denotes the yield or effect of the factors.

Trend(T), Seasonal (S), Cycle (C) and Irregular variation is (I)

Y=T*S*C*I

Y=T+S+C+S+I

The additive model is based on the assumption that the four components are independent. Additive models are very rarely used and it is not appropriate for future events.

Discrimination analysis in R

Conclusion

The essential requirements are

1. Data should be available for a long period
2. The values should be in an equal interval of a time otherwise need to adjust
3. Data should consist of a homogeneous value