Cluster Analysis on the Development of Agricultural Economy in the Cities (Zone) of Shaanxi

BACKGROUNDShaanxi is a province located in the west of China, under the jurisdiction of which there are ten cities and one demonstration zone.It is the most developed province in industry and agriculture in Northwest China.By geographical location, Shaanxi is divided into three parts, including Central Shaanxi Plain, Northern Shaanxi and Southern Shaanxi.Traditionally, the development conditions of agricultural economy in this province are also divided into three sections by geographical location, including the conditions in Central Shaanxi Plain, Northern Shaanxi and Southern Shaanxi.However, this paper, no longer following the traditional division method, divides the development conditions of agricultural economy in the cities (zone) of Shaanxi according to development indicators of agricultural economy.Through clustering and relevant analysis, it investigates the development conditions of agricultural economy in the cities (zone) of Shaanxi. EVALUATION INDICATOR SYSTEM AND EVALUATION METHOD Establishment of Evaluation Indicator SystemAt present, there are no unanimous views in the academic circle as to the method for evaluation of the development level of agricultural economy and construction of the indicator system.On the basis of a study on the indicators of various models, according to the requirement that the indicator evaluation system should have comprehensive coverage, reasonable structure, proper simplicity and complexity and strong representativeness, and following the scientific, objective, systematic, feasible and comparable principles, the author sets the agricultural development levels of the cities (zone) in Shaanxi as an evaluation indicator system consisting of indicators at three levels.These indicators include one Level I indicator (i.e., the agricultural development level), three Level II indicators, and nine Level III indicators.See Table 1 for details of the indicators. Data Collection and ProcessingThe original data used in this paper are all from Shaanxi Statistical Yearbook 2014, and partial data are obtained by calculation and sorting based on the original data.Since these data have different dimensions and orders of magnitude, they cannot be directly used for evaluation.In order to ensure the reliability of results, it is necessary to normalize the original indicator data.In this paper, the technique of z-score normalization is applied, by which the original data values are converted to new data, i.e., new data = (original datamean value)/standard deviation.