DOF ranks analysts based on forecasts

Economists and analysts from private institutions provided better gross domestic product (GDP) growth forecasts in the fourth quarter of 2018 compared to their projections in the previous quarters, according to the Department of Finance.

The DOF, through its Strategy, Economics and Results Group (SERG), analyzed the forecasts announced by 10 analysts from private institutions

for the fourth quarter of 2018 and compared them to the actual GDP growth figure of 6.1 percent during the period.

Based on the results, Finance Undersecretary Karl Kendrick Chua said the most accurate forecasts in the fourth quarter of 2018 deviated only 0.1 percentage point from the official GDP growth figure, representing an error of 1.6 percent.

On the other hand, Chua said the least accurate forecasts had a difference of 0.4 percentage point from the actual growth figure, an error of 6.6 percent.

According to the DOF official, only two out of 10 forecasts last quarter had errors greater than five percent, an improvement from seven out of 15 forecasts in the third quarter of last year.

Meanwhile, Chua said DOFSERG also made an analysis on the performance of 20 economists and analysts from institutions that regularly announced their forecasts in Manila’s daily newspapers from 2016 to 2018.

He said quarterly forecasts during the two-year period had errors averaging 5.6 percent, with 13 out of 20 economists scoring an average error of more than five percent.

“The best forecaster registered an average error of 3.1 percent, while the widest margin was 7.2 percent during the period 2016 to 2018. An error of five percent or less is considered a good forecast,” the DOF said.

Chua said SERG used the mean absolute percentage error (MAPE) method to analyze the GDP forecast errors.

“The error is defined as actual or observed value minus the forecasted value, as a percentage of the actual value. The smaller the MAPE, the better the forecasts,” Chua said.

“As with inflation forecasts, we did the assessment to see how well analysts were in forecasting GDP growth. The results, based on published forecasts from 2016 to 2018, show that accuracy, on average, has improved recently. For example, the average error in the fourth quarter of 2018 was down to four percent, from 10.1 and 5.4 percent in the second and third quarters, respectively,” Chua said.