The problem of the extraction of the relevant information for pre- diction purposes in a Big Data time series context is tackled. This issue is especially crucial when the forecasting activity involves macroeconomic time series, i.e. when one is mostly interested in finding leading variables and, at the same time, avoiding overfitted model structures. Unfortunately, the use of big data can cause dangerous overparametrization phenomena in the enter- tained models. In addition, two other drawbacks should be considered: firstly, humandriven handling of big data on a case-by-case basis is an impractical (and generally not viable) option and secondly, focusing solely on the raw time series might lead to suboptimal results. The presented approach deals with these problems using a twofold strategy: i) it expands the data in time scale domain, in the attempt to increase the likelihood of giving emphasis to possibly weak, relevant, signals and ii) carries out a multi-step dimension reduction procedure. The latter task is done by means of crosscorrelation functions (whose employment will be theoretically justified) and a suitable objective function.
Livio Fenga ISTAT, Italian National Institute of Statistics, Italy.
Introduction: An inclusive model of economical-social interactions and its repercussions on Big Data analysis is presented. Many phenomenological topics are involved in this job, such as the idea of complexity, statistical human behavior and market structures. Complexity on social interactions is a polemic subject, and it is also a complicated phenomenon to deal with.
/ Objectives: This particular study is aimed to
develop some proper mathematical model to justify the big data consuming
economical framework with the proper social interactions. So that it can build
some major key processes assessing several types of economical frames.
Design: Chain Phenomena Analysis.
and Duration of Study: University of Guadalajara, Physics
Department, Data Science Group.
This study shows how, as long as time change currently, social interaction
impact on economical framework has become bigger. Big Data tools to manipulate
high volume levels of information from these interactions have been a strongest
platform to analyse economical indicators, such as those which repercussions
affects financial stock markets. This process is modelled in this article. A
complete model could be a model that considers as social interactions more factors
than just trading. Although we are working to applied this model and test it
with concrete data bases and real examples. This set of scalar constants
defined with κ might have empty entries
waiting to be filled with experimental data and empirical tuning. Stock Markets
one have a better option about predictions then a social interaction based
model. We are not saying that this model puts aside regular methods for stock
market predictions, but, perhaps this new approach would helps to this purpose.
A. Alatorre Department of Physics, University of Guadalajara, Olimpica Boulevard General Marcelino Garca Barragn 1421, Olmpica, 44430 San Pedro, Tlaquepaque, Jal, Mexico.