题目 ： An assessment on the quality of China's preliminary data of quarterly GDP announcements

    作者 ： Yang, Lixiong （兰州大学 /Lanzhou University ）

    期刊 ： APPLIED ECONOMICS（影响因子 0.648）

                Volume 49, 期 : 54, Pages 5558-5569

    英文摘要 ：

    This article examines the quality of China’s preliminary announcements of the quarterly GDP. We modify the tests for unbiasedness and efficiency by incorporating the Fourier approximation to capture the effect of the state of the economy, and employing the Kiefer, Vogelsang and Bunzel (KVB) approach, developed by KVB in 2000 to reconstruct the tests to improve the finite sample properties. The results show that: (1) there is no enough evidence to support that the preliminary and first revised data are unbiased and efficient; (2) there exist systematic errors related to the state of the economy, and hence information about the state of the economy was not incorporated into the GDP data. Furthermore, we find that there is a possibility that these systematic errors associated with the stages of the business cycle may offset each other, and there is also a possibility that there exist offsetting errors in the underlying components of GDP.

    中文摘要 ：

    全球各经济体的 GDP 数据核算大致都分为：初步核算、初步核实和最终核实数据。初步核算数据发布最及时，但相对不准确；最终核实数据最准确，却严重滞后。因此，数据用户必须依赖初步核算数据判断当期经济运行情况，因而，评估初步核算数据的质量具有重要意义。近期的研究表明：全球主要经济体（如美国）的 GDP 初步数据都依赖于经济周期。本文评估我国 GDP 的数据质量与经济周期的关系，使用傅里叶变换捕捉经济周期的影响改进了现有的有效性检验和无偏性检验方法。研究发现：我国的 GDP 初步核算数据也存在依赖于经济周期的误差，即考虑经济周期的信息有助于促进 GDP 初步核算数据的质量。进一步的分析表明：在经济周期的不同阶段，数据的误差会相互抵消，从而使得传统的方法无法捕捉到数据误差。

DOI: 10.1080/00036846.2017.1313950