17. Antkiewicz A, Whalley J. China`s new regional trade agreements. World Econ. (2005) 28:1539-57. doi: 10.1111/j.1467-9701.2005.00746.x Second, we are interested in the evolution of annual IT values after the effective date of an agreement. statistically characterizing their trend over the interval [TItf ,..,TItf-5] (including the year the BTA was implemented and the next five years). To assess this trend, we examine two possible models: in the first, we perform a simple linear regression, with the exception of common characteristics such as the total volume of trade or the absolute values of imports or exports studied in previous factories [4, 5], IT also records indirect trade effects. These indirect effects occur, for example. B when a customer increases production and probably requires more inputs needed to produce their products. This increase in demand in turn affects the activities of the input subcontracting industry.

Ti individually covers both demand and supply effects and can be defined in any direction, distinguishing between IT input and IT output from one country to another as a business partner. The IT input (output TI) thus quantifies the relative importance of a C1 country as a supplier (consumer) for the production of another C2 country. Note that the relative economic relevance of countries is not symmetrical to each other. Moreover, ti is a relative measure in that it is based on the fraction of all business activities of a country identified by a particular partner. In a global environment where all countries increase their international exports, IT declines between two countries when the growth in bilateral trade volume is below the global average growth (since the country`s sectoral structure remains the same). Here we appreciate IT for THE ITN for every year between 1990 and 2013. with the Heaviside function, the break point (unknown) ψ, the i trend settings (i 1, 2) and a notion of gauss s (t) error as in the linear model. Unlike linear regression, the model in the equation (5) may also represent a local extreme value during the study period, which would be represented by a change in the signs of inclination between the two segments. More complex regression models at multiple breakpoints cannot be reliably used because of the approximate (annual) resolution of the data under consideration. Therefore, we do not look at these broader models and emphasize that we are only interested in the warning signs and statistical relevance of short-term (pluriles) trends after the implementation of the ATO, rather than accurately and functionally describing the form of these trends or explicit quantitative estimates.