Forecasting australian macroeconomic variables using a large dataset
Citations
6 citations
Cites background or result from "Forecasting australian macroeconomi..."
...There is little literature providing estimates of for Australia, which governs the mobility of labour between sectors in response to real-wage differentials, and as we estimate the model without using labour market data we set ¼ 0:75.9 The depreciation rates are chosen with reference to annual national accounts data, which suggest that the depreciation rate in the mining sector is lower than in the rest of the economy at nearly 5 per cent per annum....
[...]
...For Australia, Tsiaplias and Chua (2010) study several factor-based forecasts and use a BVAR as a benchmark....
[...]
...26Abbas et al. (2016) also are critical of the New Keynesian Phillips curve as a model of inflation for Australia, although they focus on evaluating variants of the curve in Galí and Monacelli (2005), rather than the approach in this model which allows for incomplete short-run pass through in import prices, which is a common specification in empirical small-open economy DSGE models....
[...]
...Given the informational advantage that the FAVAR model has, it is somewhat surprising that it does not perform better than the autoregressive or SOE Minnesota VAR for output, although this aligns with the factor forecasts of Tsiaplias and Chua (2010) and the findings of Jiang et al. (2017)....
[...]
1 citations
Cites methods from "Forecasting australian macroeconomi..."
...A relatively recent study is Tsiaplias & Chua (2010) which examines the use of large data sets using the same approach as Stock & Watson (2002a,b)....
[...]
References
705 citations
688 citations
"Forecasting australian macroeconomi..." refers background or methods in this paper
...The weighted PCA models tend to produce better forecasts than their unweighted PCA variants as h increases, with the weighted PCA models producing significantly smaller RMS errors at h = 4 for inflation and unemployment and similar RMS levels for real and nominal GDP. Forni et al. (2005) and Boivin and Ng (2006) also find that forecasts from weighted factors tend to outperform forecasts from their unweighted counterparts....
[...]
...The results differ, however, for the three-quarter and year ahead forecasts with the various factor approaches resulting in markedly different RMS errors The RMS errors for the factor models are clearly smaller than their VAR and BVAR counterparts for nominal GDP....
[...]
...…PCA models producing significantly smaller RMS errors at h = 4 for inflation and unemployment and similar RMS levels for real and nominal GDP. Forni et al. (2005) and Boivin and Ng (2006) also find that forecasts from weighted factors tend to outperform forecasts from their unweighted…...
[...]
...We fail to observe such a restriction here, instead finding that the RMSE gain associated with weighted PCA is greater for nominal GDP than real GDP....
[...]
...Similarly, Forni et al. (2005) find that factor-based forecasts performed better than AR forecasts for European inflation and production....
[...]
594 citations
520 citations
268 citations
"Forecasting australian macroeconomi..." refers result in this paper
...yi;t+h = �i;h + � 0 i;h (L)Yt + ei;t+h ( 4 ) where Yt = [y1;t;y2;t;y3;t;y4;t] 0 . A normal error distribution is assumed and maxi-...
[...]
...As such, the results appear consistent with simulation evidence that forecasts of yt using its own lags are less accurate than forecasts depending on lags of Ft (Boivin & Ng, 2005)....
[...]
Related Papers (5)
Frequently Asked Questions (5)
Q2. What are the future works mentioned in the paper "Forecasting australian macroeconomic variables using a large dataset" ?
17 To improve the forecasting performance of the di¤usion index approach, further research is needed regarding the construction of straightforward methods for ltering large datasets in a manner conducive to obtaining forecast improvements.
Q3. Why are the multivariate and AR approaches chosen as popular forecasting tools?
The AR and multivariate approaches are chosen as popular single equation forecasting methods, while the VAR and BVAR methods are chosen due to their popularity as macroeconomic forecasting tools.
Q4. Why are some variables omitted from the Australian Treasury forecast?
Some variables used by the Australian Treasury have been omitted due to their inappropriateness for forecasting purposes (such as dummy or trend variables) or their perfect correlation with existing variables in the set.
Q5. What is the reason for the weaker performance of the diusion index approach?
A possible reason for the weaker performance of the di¤usion index approach in this paper may be the number of time-dependent observations, since both the factor estimates F and the loadings are sensitive to T .