Abstract
Forecasting Turkey’s GDP Growth with Mixed Data Sampling (MIDAS) Method
Referring to the monetary value of the final goods and services produced in an economy within a country’s borders in a certain period, Gross Domestic Product (GDP), is one of the macroeconomic indicators that have been studied and its forecast to be modeled in the economics literature for many years. This study aims to forecast Turkey's GDP growth, using monthly tax revenues, in the period of 2006.Q1 - 2018.Q3 with Mixed Data Sampling (MIDAS) approach. The MIDAS methodology, eliminating the need to have the same frequency of dependent and independent variables which is a must in the traditional regression approach, allows the use of different frequency data in the same model. For this purpose, the aggregated-regression model and alternative MIDAS models were estimated and their forecasts were performed. Considering the performance of forecast models, it seems that among the MIDAS models the PGM-Almon model provides better results in forecasting GDP growth of Turkey's economy.
Keywords
GDP, tax revenues, MIDAS, mixed data sampling, Turkey.
