窦同宇, 夏安桃,胡耀躲

(湖南师范大学资源环境科学学院， 中国 长沙 410081)

基于非参数分析方法的湖南省县域经济时空演变分析

窦同宇, 夏安桃,胡耀躲

(湖南师范大学资源环境科学学院， 中国 长沙 410081)

采用核密度分析、马尔科夫链、空间马尔科夫链等非参数分析方法，分析湖南省县域经济增长的时空演变规律.研究表明：(1)整体呈现出“总体趋向分异，板块趋向收敛”的空间转移变化格局.(2)密度分布曲线经历了先向右偏移后向左偏移的转移变化；核密度值(0.6～0.8)集中区，人均GDP不断降低，表明低状态水平的县区相对更加“落后”；核密度极值区则由“双核”演变到“单核”，形成了一个高密度值的收敛区域.(3)湖南省4个状态水平的稳定性是高状态>低状态>中高状态>中低状态，在由低状态到高状态的4个区域背景下，低一级状态向高一级状态转移的概率之和由小变大.最后，根据不同状态县域经济发展的水平，提出了针对性的政策建议.

县域经济；核密度分析；马尔科夫链；空间马尔科夫链；时空演变

AbstractThis article makes use of kernel density analysis, Markov chains, spatial Markov chains and other non-parametric analysis tools to investigate the time-space evolution of county economic growth in Hunan province. Our research results show that: (1) A spatial pattern of shift changes is apparent, exhibiting that the general trend is different but the block trend is convergent. (2) The density distribution curve has undergone shift changes from moving to the right and then to the left. The kernel density value is concentrated (0.6～0.8), per capita GDP keeps falling, the county economy at low levels is relatively more “backward”, and the kernel density maximum area have evolved from the “double core” mode into the “single core” mode, so it forms a convergent region of high density values. (3) The stability order of the four states is as follows: high state > low state > medium high state > medium low state, in Hunan province. From the low state to the high state of the four-regional background, the sum of the probabilities of the transfer increased from the lower state to the higher state. Finally, according to different levels of economic development at the end of this paper, some suggestions and advice have been proposed.

Keywordscounty economy; kernel density analysis; Markov chains; spatial Markov chains; time-space evolution

区域经济的趋同性和差异性作为区域经济时空演变研究的重要内容，自20世纪90年代以来便备受关注[1-3].概括来说，区域经济增长趋同与差异的研究维度上，形成了时间和空间2个维度[4]，其中，时间维度上多采用参数分析，空间维度则更加注重非参数分析.目前区域经济增长趋同与差异的实证分析研究多集中于时间维度上的参数分析，即利用不同指标构建指标体系或运用相关性……