Szeregi czasowe jako struktura danych opisująca zmienność w czasie. Wyzwania, perspektywy i komputerowo wspomagane metody badawcze

The article presents the issue of the use of time series in historical research. It presents the specifics of this type of data, its use in economic history research and the impact of historical policy actions on historical knowledge acquisition practices (internet searches). Part of the article is a presentation of the results of research, which uses data on the history of Internet searches shown in the form of a time series. Two methods of time series analysis are shown to work on the example of this research. The first method presented is the study of the causal moment – this method consists in isolating in a time series the moment of a potential change of a trend (as a critical moment) and generating an alternative time series, which is a hypothetical extension of the trend before the causal moment) of the critical moment. The study involves comparing the real time series with the hypothetical one. Their divergence indicates a significant change taking place at the causal moment. The second method consists in extracting from the real time series its components: a linear trend, a seasonal or cyclical trend and random elements (residuals). The purpose of this study is to find out the specifics of the process – to what extent is it predictable, repeatable or chaotic?