Related variety and Unrelated Variety: NUTS-3 Türkiye

I present a new dataset for the related and unrelated variety at the provincial level in Turkey by employing Hidalgo, Klinger, Barabasi, and Hausmann's (2007) proximity approach following Boschma, Minondo, and Navarro (2012 and 2013). The data covers the period between 2007 and 2022.I used province-level export data obtained from the Turkish Statistical Institute (TurkStat) classified according to the four-digit Standard International Trade Classification (SITC). Global proximity values for product groups, on the other hand, were calculated by using the International Trade Database at the Product Level (BACI) dataset. First, the HS classification of BACI has been converted to four-digit SITC (revision 3) by using Eurostat Metadata Server (RAMON) conversion tables to provide data compliance.In addition, to eliminate the noise resulting from the fluctuations in prices, exchange rates, and seasonality, especially in four-digit export data, Hidalgo (2021) recommends excluding countries that export less than 1 Billion dollars, countries with a population of less than 1 Million, and products with a global export value below 500 Million dollars from the sample. After the data cleaning process, we calculated global proximity values over 139 countries.Variables descriptionColumn A: yearColumn B: Province namesColumn C: Province idColumn D: Related Variety IndexColumn E: Unrelated Variety Index

本文构建了一套全新的土耳其省级层面相关多样化与非相关多样化数据集，研究遵循Boschma、Minondo与Navarro（2012、2013）的研究范式，采用Hidalgo、Klinger、Barabasi及Hausmann（2007）提出的邻近性测算框架。数据集覆盖2007年至2022年的时间区间。
本研究使用土耳其统计署（TurkStat）提供的省级出口贸易数据，该数据按四位标准国际贸易分类（Standard International Trade Classification, SITC）进行分类。产品组的全球邻近性数值则依托产品层面国际贸易数据库（BACI）测算得到。
为保证数据一致性，首先借助欧盟统计局元数据服务器（RAMON）的转换对照表，将BACI数据库的协调编码制度（Harmonized System, HS）分类转换为四位SITC（第3版）分类。
此外，为消除价格、汇率波动及季节性因素带来的统计噪声，尤其针对四位码出口数据，Hidalgo（2021）建议将样本中出口额低于10亿美元的国家、人口不足100万的国家，以及全球出口额低于5亿美元的产品剔除出样本。完成数据清洗流程后，基于139个国家测算得到全球邻近性数值。
变量说明如下：
列A：年份
列B：省份名称
列C：省份编号
列D：相关多样化指数
列E：非相关多样化指数