Page 287 - DECO504_STATISTICAL_METHODS_IN_ECONOMICS_ENGLISH

Statistical Methods in Economics

Notes The above two methods of cyclical variation, per cent of trend and relative cyclical
residual, are percentages of the trend. For example, in 1976 the per cent of trend
indicated that the actual yield was 99.5 per cent of the expected yield for that year,
while for the same year, the relative cyclical residual indicated that the actual yield
was 0.5 per cent short of the expected yield during the year. It must be noted here that
the methods described above are only used for describing the past cyclical variations
and not for predicting future cyclical variations.
Use of Cyclical Variations

The following are the main uses of cyclical variations:
(a) Aid to Policy Formation: The study of cyclical variations is extremely useful in framing suitable
policies for stabilizing the level of business activity. One can avoid the periods of booms and
depressions since both are bad for an economy.
(b) Helps in Studying Fluctuations of Business: The cyclical variations are very helpful in studying
the characteristics of fluctuations of a business. One can come to know how sensitive is the
business to general cyclical influences ? The general pattern of a particular firm’s production,
profits, sales, raw material prices, etc. can also be known.
(c) Helps in Forecasting: The cyclical variations are helpful in forecasting and estimating about
the future behaviour. Accurate forecasting is a prerequisite for successful business.
(d) Knowledge of Irregular Fluctuations: The study of cyclical variations is helpful in analysing
and isolating the effects of irregular fluctuations. One can come to know either the variations
are unpredictable or are caused by other isolated special occurrences like floods, earthquakes,
strikes, wars, etc.
Seasonal Variation
Seasonal variations are those forces affecting time series that are the result of man made or physical
phenomena. The major characteristic of seasonal variations is that they are repetitive and periodic,
the period is less than one year, say a week, a month or a quarter. Seasonal variations can affect a
time during November is normal, it can examine the seasonal pattern in the previous years and get
the information it needs.
1. It is possible to establish the pattern of past changes. This helps us to compare two time intervals
that would otherwise be too dissimilar. For example, if a business house wants to know whether
the slump in sales during November is normal, it can examine the seasonal pattern in previous
years and get the information it needs.
2. Seasonal variations help us to project past patterns into the future. In the case of long range
decisions, secular trend analysis may be adequate. However, for short-run decisions, the ability
to predict seasonal fluctuations is essential. For example, consider the case of a wholesale food
dealer who wants to maintain a minimum adequate stock of all food items. The ability to predict
short-run patterns, such as the demand of food items during Diwali, or at Christmas, or during
the summer, is very useful to the management of the store.
3. Once the existence of the seasonal pattern has been established, it is possible to eliminate its
effects from the time series. This elimination helps us to calculate the cyclical variation that
takes place each year. When the effect of the seasonal variation has been eliminated from the
time series, we have deseasonalized time series.
In order to measure the seasonal variation, we use the ratio-to-moving average method. This method
provides an index that describes the degree of seasonal variation. The index is based on a mean of
100, with the degree of seasonality measured by variations away from the base.
The method of the ratio-to-moving average for computing the indices of seasonal variation is a
procedure whereby the different components in the series are measured and are isolated or eliminated.
Subsequently, the seasonal effect is identified and expressed in percentage form. We first take a
series in which seasonal pattern is suspected and plot this series on a graph to identify the recurrence

282 LOVELY PROFESSIONAL UNIVERSITY