Name: Bartolomeu Zamprogno
Type: PhD thesis
Publication date: 09/08/2013
Advisor:

Name Rolesort descending
Valdério Anselmo Reisen Advisor *

Examining board:

Name Rolesort descending
Valdério Anselmo Reisen Advisor *
Neyval Costa Reis Jr. Co advisor *
Flávio Augusto Ziegelmann External Examiner *
Wilfredo Omar Palma Manriquez External Examiner *
Taciana Toledo de Almeida Albuquerque Internal Examiner *
Jane Meri Santos Internal Examiner *

Summary: This work was motivated by the application of principal component analysis technique in
different contexts of area air pollution, especially in the use of network management. This
statistical methodology in practical terms produces information with accuracies in making
important decisions for quality air. This technique is commonly used, as well as in the
regression analysis as a tool for analysis and interpretation of the phenomena of the data.
However, according to the statistical literature that fosters basis for the use of this tool
in any area of application, the technique requires the assumption in this case the use of
independent variables, a characteristic which is hardly observed in practical situations in
the eld of air pollution. In general, the data available for troubleshooting management
network, identication of pollutant source, studies spatio-temporal association and the number
of hospitalizations for respiratory pollutants are by series displaying structure of short and long
time dependence, that is, autocorrelation. The research results show, in the eld of time that
the technique of principal components analysis, depending on the structure autocorrelation of
the series, can be based on spurious results. When the structure is weak, the autocorrelation
effect of autocorrelation is practically zero, so that the method can be used without further
problems. In the context of the use of the technique of time series analysis in the frequency
domain was reported the extension of existing methods for the case of time series data memory
long. The results show that the use of frequency domain methods can be used, but some
considerations should be observed and some types of applications, the air pollution, deserve
further study because of the difficulty of interpreting the frequency domain.
Keywords: principal component analysis, air pollution, time series analysis, time domain,
frequency domain

Access to document

Acesso à informação
Transparência Pública

© 2013 Universidade Federal do Espírito Santo. Todos os direitos reservados.
Av. Fernando Ferrari, 514 - Goiabeiras, Vitória - ES | CEP 29075-910