Forecasting Brazilian Inflation by Its Aggregate and Disaggregated Data: A Test of Predictive Power by Forecast Horizon

Abstract

This work aims to compare the forecast efficiency of different types of methodologies applied to Brazilian consumer inflation (IPCA). We will compare forecasting models using disaggregated and aggregated data from IPCA over twelve months ahead. We used IPCA in a monthly basis, over the period between January 1996 to March 2012. Out-of-sample analysis will be made through the period of January 2008 to March 2012. The disaggregated models were estimated by SARIMA and will have different levels of disaggregation from IPCA as groups and items, as well as disaggregation with more economic sense used by Brazilian Central Bank as: I) services, monitored prices, food and industrials and II) durables, non-durables, semi durables, services and monitored prices. Aggregated models will be estimated by time series techniques as SARIMA, state-space structural models and Markov-switching. The forecasting accuracy among models will be made by the selection model procedure known as Model Confidence Set (Hansen et al. (2011)). We were able to find evidence of forecast accuracy gains in models using more disaggregated rather than aggregate data.

Comparing the Accuracy of Inflation Forecasting in Brazil: A MIDAS Approach with Daily Online Prices versus Dynamic Factor Models

“…In the Brazilian literature, Arruda et al (2011) compare the forecasts of Brazilian monthly inflation rate produced by different linear and nonlinear time series and a Phillips' curve process, and observe that the VAR model including inflation and interest rates had the best predictive performance among linear models, but in general nonlinear models produced better forecasts. Carlo and Marçal (2016) compare the forecast performance of different methodologies applied to Brazilian consumer inflation over 12 months ahead and observe that the disaggregated analysis of price indices can generate more accurate forecasts, in particular for shorter forecast horizons. That is, instead of forecasting the main variable, each of its subcompenents is forecasted individually and lately reaggregated.…”

Comparing the Accuracy of Inflation Forecasting in Brazil: A MIDAS Approach with Daily Online Prices versus Dynamic Factor Models

“…In the Brazilian literature, Arruda et al (2011) compare the forecasts of Brazilian monthly inflation rate produced by different linear and nonlinear time series and a Phillips' curve process, and observe that the VAR model including inflation and interest rates had the best predictive performance among linear models, but in general nonlinear models produced better forecasts. Carlo and Marçal (2016) compare the forecast performance of different methodologies applied to Brazilian consumer inflation over 12 months ahead and observe that the disaggregated analysis of price indices can generate more accurate forecasts, in particular for shorter forecast horizons. That is, instead of forecasting the main variable, each of its subcompenents is forecasted individually and lately reaggregated.…”

Previsões macroeconômicas baseadas em modelos TVP-VAR: evidências para o Brasil

“…Entretanto, não é possível identificar uma única especificação que domine todas as demais em todos os horizontes de previsão e variáveis de interesse. Na literatura brasileira, Carlos & Marçal (2013) analisaram diferentes metodologias para a previsão da inflação (IPCA) e também encontram evidências de que não existe um modelo que supere todos os demais em todos os horizontes de previsão. Autores como R. T. Ferreira et al (2005) e Céspedes et al (2006) realizaram previsões da taxa de crescimento do PIB real do Brasil usando diferentes abordagens e também não identificaram uma abordagem que domine todas as demais.…”

Forecasting Inflation with the Phillips Curve: A Dynamic Model Averaging Approach for Brazil

“…However, little attention has been given to forecasting inflation with potential time-varying features for Brazil (see e.g. Arruda, Ferreira, &Castelar, 2011, andCarlos &Marçal, 2013). To the best of our knowledge, there is no analysis on forecasting Brazilian inflation taking into account both model and parameter uncertainty.…”

Comparing the Accuracy of Inflation Forecasting in Brazil: A MIDAS Approach with Daily Online Prices versus Dynamic Factor Models

“…In the Brazilian literature, Arruda et al (2011) compare the forecasts of Brazilian monthly inflation rate produced by different linear and nonlinear time series and a Phillips' curve process, and observe that the VAR model including inflation and interest rates had the best predictive performance among linear models, but in general nonlinear models produced better forecasts. Carlo and Marçal (2016) compare the forecast performance of different methodologies applied to Brazilian consumer inflation over 12 months ahead and observe that the disaggregated analysis of price indices can generate more accurate forecasts, in particular for shorter forecast horizons. That is, instead of forecasting the main variable, each of its subcompenents is forecasted individually and lately reaggregated.…”

Previsões macroeconômicas baseadas em modelos TVP-VAR: evidências para o Brasil

“…Entretanto, não é possível identificar uma única especificação que domine todas as demais em todos os horizontes de previsão e variáveis de interesse. Na literatura brasileira, Carlos & Marçal (2013) analisaram diferentes metodologias para a previsão da inflação (IPCA) e também encontram evidências de que não existe um modelo que supere todos os demais em todos os horizontes de previsão. Autores como R. T. Ferreira et al (2005) e Céspedes et al (2006) realizaram previsões da taxa de crescimento do PIB real do Brasil usando diferentes abordagens e também não identificaram uma abordagem que domine todas as demais.…”

Forecasting Inflation with the Phillips Curve: A Dynamic Model Averaging Approach for Brazil

“…However, little attention has been given to forecasting inflation with potential time-varying features for Brazil (see e.g. Arruda, Ferreira, &Castelar, 2011, andCarlos &Marçal, 2013). To the best of our knowledge, there is no analysis on forecasting Brazilian inflation taking into account both model and parameter uncertainty.…”

Comparing the Accuracy of Inflation Forecasting in Brazil: A MIDAS Approach with Daily Online Prices versus Dynamic Factor Models

“…In the Brazilian literature, Arruda et al (2011) compare the forecasts of Brazilian monthly inflation rate produced by different linear and nonlinear time series and a Phillips' curve process, and observe that the VAR model including inflation and interest rates had the best predictive performance among linear models, but in general nonlinear models produced better forecasts. Carlo and Marçal (2016) compare the forecast performance of different methodologies applied to Brazilian consumer inflation over 12 months ahead and observe that the disaggregated analysis of price indices can generate more accurate forecasts, in particular for shorter forecast horizons. That is, instead of forecasting the main variable, each of its subcompenents is forecasted individually and lately reaggregated.…”

Previsões macroeconômicas baseadas em modelos TVP-VAR: evidências para o Brasil

“…Entretanto, não é possível identificar uma única especificação que domine todas as demais em todos os horizontes de previsão e variáveis de interesse. Na literatura brasileira, Carlos & Marçal (2013) analisaram diferentes metodologias para a previsão da inflação (IPCA) e também encontram evidências de que não existe um modelo que supere todos os demais em todos os horizontes de previsão. Autores como R. T. Ferreira et al (2005) e Céspedes et al (2006) realizaram previsões da taxa de crescimento do PIB real do Brasil usando diferentes abordagens e também não identificaram uma abordagem que domine todas as demais.…”

Forecasting Inflation with the Phillips Curve: A Dynamic Model Averaging Approach for Brazil

“…However, little attention has been given to forecasting inflation with potential time-varying features for Brazil (see e.g. Arruda, Ferreira, &Castelar, 2011, andCarlos &Marçal, 2013). To the best of our knowledge, there is no analysis on forecasting Brazilian inflation taking into account both model and parameter uncertainty.…”