Submitted by Greta Butvilaite (butvilaite.g@gmail.com) on 2016-03-06T16:40:31Z No. of bitstreams: 1 Master Thesis. version for export.pdf: 4005967 bytes, checksum: 7182562ae47f08ac6216e5c6c4401f03 (MD5) / Approved for entry into archive by Janete de Oliveira Feitosa (janete.feitosa@fgv.br) on 2016-08-01T14:55:42Z (GMT) No. of bitstreams: 1 Master Thesis. version for export.pdf: 4005967 bytes, checksum: 7182562ae47f08ac6216e5c6c4401f03 (MD5) / Approved for entry into archive by Maria Almeida (maria.socorro@fgv.br) on 2016-08-01T19:43:34Z (GMT) No. of bitstreams: 1 Master Thesis. version for export.pdf: 4005967 bytes, checksum: 7182562ae47f08ac6216e5c6c4401f03 (MD5) / Made available in DSpace on 2016-08-01T19:43:49Z (GMT). No. of bitstreams: 1 Master Thesis. version for export.pdf: 4005967 bytes, checksum: 7182562ae47f08ac6216e5c6c4401f03 (MD5) Previous issue date: 2016-01-29 / This thesis is an analytical analysis of consumption in Brazil, based on data from the Consumer Expenditure Survey, years 2008 to 2009, collected by the Brazilian Institute of Geography and Statistics. The main aim of the thesis was to identify differences and similarities in consumption among Brazilian households, and estimate the importance of demographic and geographic characteristics. Initially, households belonging to different social classes and geographical regions were compared based on their consumption. For further insights, two cluster analyses were conducted. Firstly, households were grouped according to the absolute values of expenditures. Five clusters were discovered; cluster membership showed larger spending in all of the expense categories for households having higher income, and a substantial association with particular demographic variables, including as region, neighborhood, race and education. Secondly, cluster analysis was performed on proportionate distribution of total spending by every household. Five groups of households were revealed: Basic Consumers, the largest group that spends only on fundamental goods, Limited Spenders, which additionally purchase alcohol, tobacco, literature and telecommunication technologies, Mainstream Buyers, characterized by spending on clothing, personal care, entertainment and transport, Advanced Consumers, which have high relative expenses on financial and legal services, healthcare and education, and Exclusive Spenders, households distinguished by spending on vehicles, real estate and travelling.

Improving the energy efficiency of buildings is important for many reasons. There is a large body of data detailing the hourly energy consumption of buildings. This work studies a large data set from the Swedish energy market. This thesis proposes a data analysis methodology for identifying abnormal consumption patterns using two steps of clustering. First, typical weekly energy usage profiles are extracted from each building by clustering week-long segments of the building’s lifetime consumption, and by extracting the medoids of the clusters. Second, all the typical weekly energyusage profiles are clustered using agglomerative hierarchical clustering. Large clusters are assumed to contain normal consumption pattens, and small clusters are assumed to have abnormal patterns. Buildings with a large presence in small clusters are said to be abnormal, and vice versa. The method employs Dynamic Time Warping distance for dissimilarity measure. Using a set of 160 buildings, manually classified by domain experts, this thesis shows that the mean abnormality-score is higher for abnormal buildings compared to normal buildings with p ≈ 0.0036.