Modelling of Joint Dependence FTSE 100 Index and Investor's Activity Level on the Internet

This work is devoted to modeling capabilities depending on the national index of the UK FTSE 100 index and information indicies of the interest and sentiment. The aim of this work is to analyze and, if possible, calculate dependency models of the national index of the UK FTSE 100 index and investor's activity on the Internet. The relationship of the closing prices of the FTSE 100 index and information indicies was analyzed. Data were uploaded from sources that include Bloomberg, Yahoo! Finance, as well as Google database and Twitter. To extract data from Google and Twitter parser-code was used. Attempts were made to construct models for the absolute values, and for returns. Analysis of the data structure showed that user sentiment index appeared to be stationary, so the attempt was made to construct a model of vector autoregression for the largest and most well-known companies included in the index. To see how the sentiment indices and interest of users affect the average yield, we attempt to evaluate the ARMA model . The analysis was based MA (1) model, in which the mood index was significant, while as the index of interest was excluded from the model due to insignificance.