Announcement

Dear Users,
Currently, all functions of the GTI system except the Report Center have been fully restored. The Report Center remains under maintenance and is expected to resume on September 8, 2025.
If you notice any missing Customized HS Group, Country Group, or Region Group in the Customize Console, please contact us via Contact Us for assistance in quickly restoring your customized groups.
We sincerely apologize for the inconvenience caused by the maintenance work over the past few days.
Thank you for your understanding and trust.
SINOIMEX - Global Trade and Industry Growth Lab

Do not remind me

功能更新提醒（2024年12月24日）

增加按表内值横纵排序的功能

完成查询后前端显示数据项的右侧新增排序小图标，点击即可对该数据项（行或列）进行排序（点击一次降序，点击两次升序）。需注意，排序仅对前端显示数据生效，超出限制条数的不参与排序。该排序可对任意数据项生效，无论行列与制表结构。例如，点击“年份”则可对所选年份本身进行升降序排序，点击单个年份则对该年下属数据项（例如进出口额）进行升降序排序，点击月份则对其所属年份下（注意并非全局）的月进行排序。
一次只能对一个同层级的数据项进行排序，整个表的结构也会随之变化。可以同时对不同层级的数据项进行多次排序（如年份-该年内每月出口额）。排序对分析范围内（如年-月）的数据生效。

增加灵活调整前端显示数据条数功能

当未选择多重分解触发下载条数限制，且前端显示数据大于3000条时，可点击“添加数据”来自行决定前端显示数据条数：1000-100000条不等（可进行手动输入），仅在当前浏览器设置生效。需注意，条数过多会影响加载速度，请根据自己的设备配置水平进行调整。

去内贸选项与指数占比运算关联

报告国组去内贸选项现可影响指数运算中的有关占比，例如，“东亚对全球海产品出口”查询中，若开启去内贸，则“占报告国份额”为将东亚视作整体但完成去内贸运算后的占比。也即占比运算的分母部分会实时受到去内贸运算的影响。

数据库更新逻辑与节奏变动

国际贸易统计数据库的更新逻辑发生了变化：基于对集群数据库的优化，国际贸易统计数据库现已实现实时更新，而非仅月度更新（镜像数据仍维持原有更新节奏）。三个数据库的更新逻辑现已统一并基于同一个集群数据库进行同步更新。该更新会一并影响所有使用该数据库的功能模块：可视化实验室-探索全球贸易、数据中心-国际贸易统计数据库、可视化实验室-探索SITC编码下的全球贸易、数据中心-贸易预测、产品空间（更新至2023年）、供应链流向地图、全球潜力市场。

数据字典表上方的解释

在数据字典页面的顶部新增了对国际贸易统计数据库、国家原始数据库、标准化省州数据库所对应三个字典的说明说明涉及三个方面：时间意义上的更新进度（哪些数据更新到什么日期）、该数据库所应用在的模块（例如可视化实验室）、数据特点（对数据本身的释义及用途解释）。

基于数据库更新的查询重构

因为可视化实验室的数据源为国际贸易统计数据库，因此随着该数据库的更新，可视化实验室也一并进行了更新：由每月整体打包更新上传数据，变为随时上传已更新的数据。该更新效果可在全球贸易总额可视化中体现：如图，现可查询上个月的数据（更新中），并随着时间推移逐渐完善。

Do not remind me

Global Import & Export Trade Forecasting System

The GTI Global Import & Export Trade Forecasting System is designed to provide users with predictive data for over 200 countries and regions worldwide, covering up to 5 years into the future. It includes three sub-modules: Reporter Country, Reporter Country's Products, and Reporter Country's Trade Partners. Within each sub-module, users can analyze and forecast trade values or quantities for imports, exports, and overall trade, with results displayed in multi-dimensional line charts and data tables. Users can select the reporter country, product (HS2/HS4/HS6), trade partner, forecast period, statistical method, and trade flow direction to generate the desired forecast.

Forecast Reporter
Forecast Reporter Product
Forecast Partner
Custom Forecast

Statistical Method

Forecasting Algorithm Introduction✕

Trade data is typically time-series data. However, relying on a single time-series forecasting algorithm may not provide the most accurate predictions. To improve forecasting performance, our prediction module integrates time-series forecasting algorithms with machine learning techniques.

First, after obtaining the historical dataset for a specific forecast condition, we perform trend, seasonal, and cyclical analyses on the data. This helps to identify the trend patterns, seasonal patterns, and cycles, and to define the range of parameters to be included in the model. (The historical dataset refers to the data from the GTI International Trade Statistics Database.)

Next, we use Optuna, a tool for hyperparameter optimization, to fine-tune the integrated model for optimal performance. We evaluate the model's performance using Root Mean Square Error (RMSE) as the metric. To ensure more accurate predictions, our training set includes at least 60 months of historical data, with a validation set consisting of at least 5 months of data.

Finally, the trained model is used to predict data for the time range selected by the user.