Estimation of Korean Monthly GDP with Mixed-Frequency Data using an Unobserved Component Error Correction Model(EP Vol.11 No.1)

Author: Ki-Ho Kim

Since GDP is announced on a quarterly basis, the data frequency of GDP
is relatively lower than that of other major macro variables published on a
monthly basis. Therefore, in actual fact, the utility of GDP in judging the
economy swiftly is not great. In terms of data utility, its utility is also low. If
GDP is used, the other monthly data must be converted into quarterly data,
and the result is a reduction to one-third of the total number of observations.
In this paper, monthly GDP is obtained by using monthly information data
such as the IPI (Industrial Production Index) and the WRSI (Wholesale and
Retail Sale Index), which are closely connected with GDP. To acquire
monthly GDP, this paper suggests applying an unobserved component vector
error correction model that can utilize mixed frequency data in estimation
directly, without extra procedures for data transformation (e.g., monthly to
quarterly).
In order to judge whether monthly GDP is estimated appropriately, this
paper compares actual quarterly GDP with converted quarterly GDP based
on the estimated monthly GDP. The RMSPE (root mean square percentage
error) of this estimated quarterly GDP shows around 0.01. The final monthly
GDP is produced by combining two individual monthly GDP figures (one
estimated using the IPI and the other by the WRSI) and then benchmarking
the combined figure, and it is then compared to the actual quarterly GDP.
The RMSPE of the final monthly GDP is close to 0. Meanwhile, the
correlation coefficient and the lagged correlation coefficient between the
final monthly GDP and the cyclical component of the CCI (Coincident
Composite Index) are high and significant, and the final monthly GDP leads
CCI by approximately one to three months.