A carregar...
Publicação
Analyzing the New Customized Monetary Unit Sampling Method
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 1.97 MB | Adobe PDF |
Orientador(es)
Resumo(s)
This work aims to analyze the New Customized Monetary Unit Sampling method (Standard
MUS) and establish a comparison with the Conservative approach. The latter is a
rudimentary method lacking detailed formulas, especially regarding sample size and
precision estimation. This limitation leads to large sample sizes, which is an undesired
outcome, both in statistics and auditing contexts, as it may potentially increase the sampling
costs or be more time-consuming. Therefore, the development of the new method
addresses the identified disadvantages. It is similar to probability proportional to size
sampling and performs well when book values have high variability and a positive correlation
exists between book values and errors. This thesis addresses the research gap concerning
the conditions under which the Standard approach outperforms the Conservative MUS. The
primary objective is to compare the two methods in terms of the estimators produced, their
accuracy, and precision. To achieve this, a Monte Carlo simulation was conducted using a
simulated population to evaluate the statistical properties of each method. The results
showed that both approaches yielded approximately unbiased estimators of the total error,
with similar variances. However, significant differences arose in precision estimation: the
Standard approach exhibited a higher empirical precision (low relative bias of –3.85%). In
comparison, the Conservative method overestimated precision (relative bias of 331.76%).
These findings suggest that the Standard MUS offers a more efficient alternative when the
population displays high variability and a positive correlation between book values and
errors, as expected.
Descrição
Dissertation presented as the partial requirement for obtaining a Master's degree in Statistics and Information Management, specialization in Information Analysis and Management
Palavras-chave
MUS Monte Carlo Simulation Sampling Methods Auditing SDG 8 - Decent work and economic growth SDG 16 - Peace, justice and strong institutions