Patterns and trends in data: Trend and pattern visualization in charts
Pattern and trend are defined as “any regularly repeated arrangement, especially a design made from lines, shapes, or colors on a surface, and the trend is the general direction of changes or developments” by the Cambridge Dictionary. The patterns and trends in data reveal important evidence and information. The organization can use this data for forecasting and planning, as well as testing theories and strategies. The methods of trend and pattern analysis are discussed as follows.
Linear Trend or Secular Trend
A linear pattern represents a constant decrease or growth in numbers over time. This data shows up or down in a diagram as a straight line (the angle may be steep or shallow). The trend can either rise or fall. The term ‘secular’ refers to the term ‘long-term’ or to ‘long-term periods.’ The secular trend refers therefore to the movement of a time series over a rather long period in one direction. Movement in nature is smooth, consistent, and regular. Such a movement marks the overall pattern of economic or social growth or decrease.
This method generates nonlinear curved lines in which the data rises or falls at a faster rate than at a steady rate. If the trend is upward, the graph will show a curved line with the last point in later years being higher than the first year, rather than a straight line pointing diagonally up.
The line in this analysis is curved to display data values increasing or falling at first, and then stopping rising or falling at a point where the trend (increase or decrease) stops rising or falling.
These are short-term variations that a time series usually follows during corresponding months or seasons of successive years. It may be due to some variation of repeating nature, triggered by repetitive events, over the course of a year. For example, increased demand for woolen clothing during the winter increased department store sales before Eid, increased candy sales before Christmas, and so on.
Long-term fluctuations or swings around the trend line or curve are referred to as cyclical movements. Cycles are also used to describe movements that take the form of upward and downward swings. Only if the movements recur after more than a year are they considered cyclical. The term ‘cycle’ refers to the ‘business cycle,’ which has four phases: boom, recession, depression, and revival.
These movements refer to erratic fluctuations caused by unforeseen events such as war, flood, storm, earthquake, accidents, strikes, and so on. They are also referred to as ‘irregular,’ ‘accidental,’ or ‘erratic’ movements.
A stationary time series has statistical properties such as mean and variance that remain constant over time. A stationary series has a constant variance and varies around a constant mean level, neither decreasing nor increasing systematically over time.