Forecasting Turkish real GDP growth in a data-rich environment

Macroeconomic forecasting with mixed data sampling frequencies: Evidence from a small open economy

“…Consistent with the findings of Monteforte and Moretti (2013) and Dogan and Midilic, (2018), our results show that MIDAS regression using high-frequency stock returns data produces a better forecast of GDP growth rate than the other models, and the best forecasting performance is achieved using weekly stock returns. The challenge is how to best use available data in our economic forecast.…”

“…In this paper, we have comparatively investigated the forecasting performance of the three models-that is, the MIDAS regression model, the direct regression model on high-frequency data, and the time-averaging regression model-by using data from the Singapore economy. Consistent with the findings of Monteforte and Moretti (2013) and Dogan and Midilic, (2018), our results show that MIDAS regression using high-frequency stock returns data produces a better forecast of GDP growth rate than the other models, and the best forecasting performance is achieved using weekly stock returns. It is also found that the intra-period MIDAS model outperforms other forecasting models, as it can capture well all the important ups and downs of the economic performance in Singapore, especially during the economic crises in 2001-2002 and 2008-2009, with the lowest RMSE value.…”

“…In addition, Chow and Choy (2009) employed a factor-augmented vector autoregressive model to forecast GDP growth rates and major sectoral output indicators for Singapore. Some more recent papers employing the MIDAS framework to forecast GDP growth rates in small open economies include Chernin and Kirchner (2017) for Chile, Dogan and Midilic (2018) for Turkey, Galli, Hepenstrickz, and Scheufele (2016) for Switzerland, Gomez-Zamudie and Ibarra (2017) for Mexico, Kim and Swanson (2017) for Korea, and Yau and Hueng (2016) for Taiwan. However, this study is among the first to employ the MIDAS approach to forecasting Singapore quarterly GDP growth rates using stock returns observed at different frequencies.…”

A hybrid MIDAS approach for forecasting hotel demand using large panels of search data

“…The results show that this approach significantly improves prediction accuracy relative to the single-factor MIDAS model. Sen Dogan and Midilic (2019) successfully predicted Turkey’s economic growth by using the two complementary methods of principal component analysis and forecast combinations. Similar studies have also been conducted by other researchers (Andreou et al, 2013; Kim and Swanson, 2017; Stock and Watson, 2004).…”

“…The dynamic factor model is used to extract feature information from large panels of key word variables, which reduces the dimension of prediction variables and avoids information redundancy and multicollinearity. To ensure the parsimony of modeling and the stability of prediction, the prediction results of the single-factor MIDAS models are combined with forecast combinations (Sen Dogan and Midilic, 2019). The results show that compared with the benchmark model, the introduction of MIDAS can improve the prediction ability of the model.…”