Prediction of Baltic Sectorial Share Price Indices
Rimantas Rudzkis
Vilnius University, Lithuania
Roma Valkavičienė
Vilnius Gediminas Technical University, Lithuania
Virmantas Kvedaras
Vilnius University, Lithuania
Published 2014-12-20


sectoral indices of stock prices
OMX Baltic security market
data aggregation
autoregression analysis

How to Cite

Rudzkis R., Valkavičienė R. and Kvedaras V. (2014) “Prediction of Baltic Sectorial Share Price Indices”, Lithuanian Journal of Statistics, 53(1), pp. 53-59. doi: 10.15388/LJS.2014.13894.


Extending the research started in [31], the paper uses econometric methods for the short-term forecasting of quarterly values of sector indexes of stock prices from the OMX Baltic stock exchange. The ARMA models and modelling methodology that was used to build the statistical models in the previous paper are now augmented with the algorithms of time series aggregation and identification of special features of the series. Here, the search for informative factors relies on the study of related literature. The specification of models is further tailored using the traditional significance (p-value) analysis of regressors and a cross-validation analysis. The latter is implemented in this paper using the Jack-knife approach. The data period analysed covers the years 2000–2013. The results of the analysis indicate that the inclusion not only of recent autoregressive terms but also of some aggregated characteristics (as certain special features of indexes) improves the precision of forecasting substantially. The calculations were performed using the statistical analysis software SAS.

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Please read the Copyright Notice in Journal Policy