Banca de DEFESA: ALEX DOS SANTOS VALLE

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : ALEX DOS SANTOS VALLE
DATE: 15/07/2025
TIME: 13:00
LOCAL: Sala virtual
TITLE:

ANALYSIS AND FORECAST OF RIO DE JANEIRO STATE'S ICMS. 

KEY WORDS:

ICMS; Time Series; Rio de Janeiro; Forecast

PAGES: 35
BIG AREA: Ciências Sociais Aplicadas
AREA: Economia
SUMMARY:

The tax revenues of the State of Rio de Janeiro (SRJ) have been investigated over recent years due to their economic importance. The ICMS (Imposto sobre Operações Relativas à Circulação de Mercadorias e sobre Prestações de Serviços de Transporte Interestadual e Intermunicipal e de Comunicação) is the highest revenue-generating tax for the SRJ, and the fiscal responsibility law requires projections for each fiscal year. In this context, short-, medium-, and long-term ICMS revenue forecasts are crucial for decision-making regarding public spending, thus contributing to the SRJ's economic development. This project proposes an analysis and forecasting of SRJ's ICMS through time series and neural network models. These models will be integrated with a bottom-up approach that allows for assessing the impact of public policies, such as laws, on each economic sector of the ERJ. Specifically, the Box-Jenkins models, exponential smoothing, dynamic regression, and neural networks (NNAR, MLP, and LSTM) will be used. Preliminary results have shown that, among the Box-Jenkins, exponential smoothing, and NNAR neural network models, the seasonal ARIMA model was the most suitable for the last year analyzed; however, on average, the exponential smoothing model delivered the best forecasting results.

COMMITTEE MEMBERS:
Externo à Instituição - RODRIGO FLORA CALILI - PUC - RJ
Interna - 2829205 - DEBORA MESQUITA PIMENTEL
Presidente - 2639882 - FELIPE LEITE COELHO DA SILVA
Interna - 1863173 - MARIA VIVIANA DE FREITAS CABRAL