Are Professional Forecasters Rational? Evidence for Brazilian Dataset

We test forecast rationality for Brazilian inflation using Survey of Professional Forecasters (SPF) for each month. We consider panel data traditional tests as Mincer and Zarnowitz (1969) and West and McCracken (1998) to verify if forecast errors have zero mean and are uncorrelated with the forecasts for both databases. We also use time series traditional tests and Rossi and Sekhposyan (2016) test with their asymptotic critical values and alternatively we compare the results with finite sample adjusted distribution critical values of El-Shagi (2019). This work is the only one that uses the fluctuation rationality test with finite sample adjusted distribution critical values for forecast rationality besides El-Shagi (2019). We do not reject the null hypothesis of forecast rationality with restricted version with panel data. We reject this null hypothesis with traditional time series approach for the consensus inflation but we do not reject the null hypothesis using fluctuation rationality test of Rossi and Sekhposyan (2016). We obtain that there is bias in inflation forecasts in the easing and tightening periods of monetary policy or election periods with panel data. But we have that economic cycle and monetary policy do not affect the rationality test with panel data. The consensus forecast seems to neutralize the bias of individual forecasts when we test considering panel data and it reduces irrationality only for periods of recession, monetary policy tightening and without election.