Skip to main content
Login | Suomeksi | På svenska | In English

Browsing by Subject "time series analysis"

Sort by: Order: Results:

  • Forecasting euro area inflation : Does internet search query data improve forecast accuracy? 
    Widgrén, Miska (2018)
    Internet search engines produce large amounts of data. This thesis shows how the data about internet searches can be used for inflation forecasting. The internet search data is constructed from searches performed on Google. The sample covers eurozone countries over the period from January 2004 to July 2017. The performance of the internet searches is evaluated relative to traditional inflation forecasting benchmark models. The usefulness of the Google searches is evaluated by Granger causality and out-of-sample performance. Furthermore, to study the robustness of the results, the out-of-sample forecasting accuracy has been evaluated in two separate sub-samples. In this study, a simple autoregressive model augmented with internet searches is found to outperform the traditional benchmark models in predicting the month-over-month inflation of the near future. Moreover, the improvement is statistically significant in one-month ahead forecasting accuracy. The Google model also outperforms the benchmark models in year-over-year inflation forecasting. However, the improvement in year-over-year forecasting accuracy is modest. In addition, this thesis shows that the seasonally adjusted internet search data can improve the performance of the Google model slightly. This thesis is related to fast-growing research on employing Google Trends data in economic forecasting. The findings in this thesis require further research in exploiting the internet search data in macroeconomic forecasting.
  • Time series analysis of IM Pegasi photometry : comparing two method 
    Rantakylä, Julia (2020)
    Active longitudes are areas, where star spot activity is centered in and reappears on a periodic manner. Star spots are cooler areas on the star surface, caused by rising magnetic field lines inhibiting the flow of the convective region. The ways to observe active longitudes is limited, but in some stars the phenomenon has clearly been present, as Lehtinen et al. (2016) has showed. One of the observation methods is to analyse the primary and secondary minima epochs of the star’s light curve relative to its orbital period. Time series analyses are tools to gather these phases from light curves. Here two different methods were used to analyse a RS CVn binary member IM Pegasi. Continuous Period Search (CPS) (Lehtinen et al. 2011) defines an adaptive,single periodic model to a moving window of observations, allowing the light curve to contain sudden changes. Discrete Chi-square Method (DCM) (Jetsu 2020)) applies a multiperiod, polynomial-trended model to fit to the data with constant parameters, assuming all changes in the light curve are part of periodic changes. Using these two methods the light curve of IM Pegasi is studied in order to determine if there could be active longitudes present. Four data segments were chosen to be further analysed with DCM based on the CPS results. One of the segments showed a flip-flop effect in the CPS phase results, which was showed to be apparent based on the successful DCM performance. Two segments, which had rather steady phase trend in the CPS results, performed well with the DCM analysis. The fourth segment, which showed strong migrating of the secondary minima phase in CPS analysis, had problems performing with DCM as a whole segment. The primary periodicity is detected in both CPS and DCM withing good limits of agreement. The DCM dual-periodic model results in all four segments indicate of an additional, more fragile irregular structure in the star, like separated dynamo waves.
  • Time Series Analysis of Solar Wind Magnetic Field Fluctuations Using Information Theory and Complex Networks 
    Koikkalainen, Venla (2023)
    The aim of this study is to inspect fluctuations in the solar wind magnetic field for four different types of solar wind time series. The events considered are fast and slow solar wind, along with magnetic clouds and sheath regions, which are found in coronal mass ejections (CMEs). Time series measurements of these processes are analysed using methods from Information Theory and Complex Network Analysis. The techniques that are used here are the Fisher-Shannon information plane, the Jensen-Shannon complexity-entropy plane, and Horizontal Visibility Graph Analysis. Statistical and information theory measures as well as network analysis have recently been applied to studying time series in an attempt to determine their internal structure. There is promising research into these methods quantifying data as either chaotic, stochastic, or periodic. Knowing whether a process has e.g. a deterministic origin could shed light on the creation of said process. Applying these methods to solar wind, more information could be found about its formation at the Sun. In general, the solar wind data analysed in this thesis was found to be stochastic, which agrees with previous studies. In addition, when analysing magnetic field magnitude B, magnetic clouds appear to have more internal structure in the time series signal than the other types of solar wind data tested. The results obtained here are promising in terms of finding differences in structure within solar wind, and could be investigated further with the use of more solar wind data.

Yhteystiedot

HELSINGIN YLIOPISTO