Некоторые результаты, касающиеся эффекта сезонной корректировки данных в динамических моделях
The paper analyses the necessity of seasonal adjustment in dynamic models. It is shown, that seasonal adjustment of the time series can influence its’ properties in terms of unit root and cointegration tests. This influence depends of the seasonal adjustment procedure and the test selected. If there is a cointegration between series, seasonal adjustment of any type reduces the quality of estimates of parameters of cointegration equation if the seasonality in original series is such that there is no seasonality in cointegration equation. If seasonality is present in the cointegration equation, seasonal adjustment increases the quality of estimates and identification of the presence of cointegration.
Seasonality and cyclicity - are two influential factors that affect dynamics of macroeconomic indicators both during the year and longer periods of time. In this article are discussed methodological questions that arise during seasonal decomposition of the GDP by factors for the year when balance aggregate and factors ratio is constant. Economic cycles mechanisms origin and their identification questions based on the combination of classical methods of spectral analysis and historic approach. Presented is the fact that along with more regular cycles such as investment and Kondratiev wave, influence of shocks (such as «oil prices crises») appear so called causal cycles that lead to a serious change in technological base of production. Particular importance (emphasis is placed on ) a new technological wave which is expected to strike the world in 2020 th and those goal set before the Russia. This research is done on the basis of world and Russian (national) statistics.
To large organizations, business intelligence (BI) promises the capability of collecting and analyzing internal and external data to generate knowledge and value, thus providing decision support at the strategic, tactical, and operational levels. BI is now impacted by the “Big Data” phenomena and the evolution of society and users. In particular, BI applications must cope with additional heterogeneous (often Web-based) sources, e.g., from social networks, blogs, competitors’, suppliers’, or distributors’ data, governmental or NGO-based analysis and papers, or from research publications. In addition, they must be able to provide their results also on mobile devices, taking into account location-based or time-based environmental data. The lectures held at the Third European Business Intelligence Summer School (eBISS), which are presented here in an extended and refined format, cover not only established BI and BPM technologies, but extend into innovative aspects that are important in this new environment and for novel applications, e.g., pattern and process mining, business semantics, Linked Open Data, and large-scale data management and analysis. Combining papers by leading researchers in the field, this volume equips the reader with the state-of-the-art background necessary for creating the future of BI. It also provides the reader with an excellent basis and many pointers for further research in this growing field.
The main goal of this paper is to study interconnections between credit ratings and financial indicators of industrial companies from BRICS countries. We use method of patterns, one of the modern methods of nonlinear modeling, to identify groups of heterogeneous objects with different influence on ratings. Additionally, in this research, we evaluate Tobit regression model for selected groups and establish some credit rating patterns for the BRICS industrial companies. Our results of Tobin model, may have practical implementation in short-term financial management.
The analysis of short-term tendency of economic dynamics can be performed on seasonally adjusted data only. This implies that each time series is to be transformed in two: the seasonal component and the remaining part. The result of such decomposition depends on the specific features of the seasonal adjustment algorithm. Most uncertainty is expected within the neighborhood of crises when the economic indicators are likely to demonstrate substantial changes. Under such circumstances, the seasonal adjustment procedures are likely to generate spurious signals that deteriorate the seasonally adjusted series.
In this paper we analyze distortions of seasonally adjusted time series of economic data that appear in the neighborhood of crises. We examined the aberrations caused by sharp level shifts as well as by changes in seasonal pattern and showed that under these circumstances the standard algorithms of seasonal adjustment can generate spurious signals similar to first signs of a crisis or its second and following waves. We consider these misleading signals from two points of view: first, as an economic historian who operates with long time series of unchanging data; second, as an analyst of short-term dynamics monitoring the data that is subject to revisions.
We show that these aberrations can be misleading for understanding of short-run dynamics especially during the first years after a crisis. The identification of the end of a recession and estimation of seasonally adjusted values of observations right after the peak (or bottom) of a fluctuation seem to be the most problematic. Monitoring within this “blind zone” appears to be very complicated. We compared aberrations produced by X-12-ARIMA and TRAMO/SEATS. Some recommendations to soften the distortions are proposed.