In any scientific field and field of knowledge there are phenomena whose study it is advisable to produce, taking into account all the changes over a given time period. As for the everyday environment of a person, it is of interest to him, for example, how prices for a particular product changed over the past year, as shown by regular examinations in medical clinics, etc.
In statistics, the totality of changes occurring with a particular object over a certain period of time is nothing more than a time series. At any level of this characteristic at a given point in time, a number of factors influence, each of which can be attributed either to random or to system-forming moments that affect both the short-term tendency and cyclic fluctuations.
Analyzing a different combination of these factors, we can conclude that the time series, depending on a particular sphere, can take one of the following forms. Firstly, a significant part of economic indicators at both the macro and micro levels is in constant dynamic change, since they are influenced by a huge number of factors. At the same time, despite the fact that these factors are often directed in different directions, in their aggregate they form a unidirectional trend, showing progress or regression in the development of a particular indicator.
Secondly, considering the time series for a particular indicator, you can clearly see that it is undergoing noticeable cyclical fluctuations. This may be due to the change of seasons, global trends or the length of the cycle of the performance of certain works.
To find out what actual characteristic a time series has at a given moment in time, it is necessary to add or multiply its random trend and cyclic components in a vector. The result obtained as a result of the addition will be an additive model of the time series, and if we apply multiplication, then in the end a multiplicative model will be presented.
The main task of any statistical study is to determine the quantitative indicators of all three main components of a given time series. This is necessary in order to predict the values of this series that can expect us in the future.
In a number of cases, scientists are required to sample a certain number of observations at approximately equal intervals of time, that is, have a stationary time series. It is obtained in those cases when the trend is removed from the dynamic time series, that is, the factors by which the formation of short-term trends occurs.
Thus, a time series is a set of quantitative values of a particular indicator taken over a certain time period. The formation of each level is influenced by many factors that are both short-term and long-term in nature.