Banca de DEFESA: Jose Carlos Ferrer Simões

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : Jose Carlos Ferrer Simões
DATE: 06/01/2023
TIME: 19:00
LOCAL: Sala de reuniões da EPR/FT
TITLE:

Evaluation of the Efficiency of Investments in Physical Security in a Financial Institution

KEY WORDS:

Data Envelopment Analysis; Operational Risk; Banking Security; Banks and Tobit Regression

PAGES: 107
BIG AREA: Ciências Exatas e da Terra
AREA: Ciência da Computação
SUMMARY:

Financial institutions invest heavily in physical and property security with the implementation of increasingly modern security devices, structural reinforcement works in treasuries and safe rooms and reinforcement of the surveillance system with the purpose of mitigating losses resulting from external thefts and fraud against agencies. However, it has been observed that the investment of resources in certain agencies, with a physical security bias, does not guarantee the same degree of relative efficiency in relation to other agencies that received less investment. In this way, an analysis of the minimum necessary investments will be carried out, with a focus on physical banking security, so that the operational risk, in the external event modality (external theft and fraud against the banking system), remains within the risk appetite established by the institution.  Note that in this study, only variables and investments that deal with physical and property security will be considered. The analysis will be in a Brazilian financial institution, through the modeling of Data Envelopment Analysis (DEA), of the investments in physical and patrimonial security, minimum necessary, in each branch, also called DMU, in order to verify the branches that present better relative efficiency compared to other agencies. For the application of DEA, 4 inputs were used (investment in physical and property security equipment, investment in infrastructure works, expenditure on surveillance and number of guards per branch) and 2 outputs (mitigated loss and inverse of the potential risk). Through this analysis, it is possible to verify the Regions, UF and DMUs that are efficient as well as to present possible reductions in security investment. It can be verified which efficient DMUs will act as a benchmark for the less efficient DMUs. Finally, the Tobit Regression was used with the purpose of verifying possible influences of indirect factors that can impact the efficiency scores.

BANKING MEMBERS:
Interno - 1114843 - ARI MELO MARIANO
Presidente - 3164711 - JOAO CARLOS FELIX SOUZA
Externo à Instituição - PENG YAOHAO - UnB