
Dynamic Double Spatial Autoregressive Model and Impulse Response Analysis
Xinxian LI, Kunpeng LI, Weiming LI
China Journal of Econometrics ›› 2021, Vol. 1 ›› Issue (1) : 66-83.
Dynamic Double Spatial Autoregressive Model and Impulse Response Analysis
Dynamic double spatial autoregressive (DDSAR) model is one of popular models in spatial econometric analysis, and is widely used in applications for its generality. This paper provides a new tool in a DDSAR model-impluse response analysis. The new tool captures the average change of dependent variable due to one-unit change of explanatory variable and the spatial structure existing in disturbances and dependent variables. Following the classical studies (LeSage and Pace (2009)), we decompose the effect into direct, indirect and total effects, and define the respective dynamic values over time and the accumulated ones. The paper provides the estimation method and inferential theory on these values. Monte Carlo simulations confirm our theoretical results and show good finite-sample performance of the estimators. We apply the new econometric tool to the regional economic development of China, and particularly investigate the nexus of human capital and GDP per capita. We find that the impulse response of indirect effects is positive in the short term and negative over the long term, implying that the relationship of Chinese regional economies are win-win in the short term and competitive over the long term. Policy suggestions are made.
dynamic models / spatial autoregressive models / maximum likelihood estimation / impulse response / economic growth {{custom_keyword}} /
表1 极大似然估计量的有限样本表现极大似然估计量的有限样本表现 |
偏倚校正之前的表现 | ||||||||||||
Bias | RMSE | SRMSE | Bias | RMSE | SRMSE | Bias | RMSE | SRMSE | ||||
50 | 50 | -0.0004 | 0.0159 | 0.7950 | -0.0057 | 0.0104 | 0.5200 | -0.0019 | 0.0130 | 0.6500 | ||
75 | 50 | -0.0003 | 0.0126 | 0.7716 | -0.0052 | 0.0088 | 0.5389 | -0.0027 | 0.0107 | 0.6552 | ||
100 | 50 | -0.0006 | 0.0111 | 0.7778 | -0.0053 | 0.0082 | 0.5798 | -0.0026 | 0.0092 | 0.6505 | ||
50 | 75 | -0.0010 | 0.0124 | 0.7593 | -0.0033 | 0.0079 | 0.4838 | -0.0014 | 0.0107 | 0.6552 | ||
75 | 75 | 0.0000 | 0.0102 | 0.7650 | -0.0038 | 0.0072 | 0.5400 | -0.0018 | 0.0089 | 0.6675 | ||
100 | 75 | 0.0001 | 0.0095 | 0.8227 | -0.0036 | 0.0063 | 0.5456 | -0.0016 | 0.0077 | 0.6668 | ||
50 | 100 | -0.0008 | 0.0114 | 0.8061 | -0.0029 | 0.0068 | 0.4808 | -0.0012 | 0.0090 | 0.6364 | ||
75 | 100 | -0.0002 | 0.0088 | 0.7621 | -0.0028 | 0.0059 | 0.5110 | -0.0012 | 0.0074 | 0.6409 | ||
100 | 100 | 0.0000 | 0.0077 | 0.7700 | -0.0025 | 0.0050 | 0.5000 | -0.0014 | 0.0067 | 0.6700 | ||
Bias | RMSE | SRMSE | Bias | RMSE | SRMSE | Bias | RMSE | SRMSE | ||||
50 | 50 | -0.0013 | 0.0109 | 0.5450 | -0.0002 | 0.0229 | 1.1450 | -0.0053 | 0.0153 | 0.7650 | ||
75 | 50 | -0.0007 | 0.0085 | 0.5205 | -0.0001 | 0.0198 | 1.2125 | -0.0053 | 0.0129 | 0.7900 | ||
100 | 50 | -0.0007 | 0.0074 | 0.5233 | 0.0000 | 0.0160 | 1.1314 | -0.0054 | 0.0115 | 0.8132 | ||
50 | 75 | -0.0009 | 0.0085 | 0.5205 | 0.0010 | 0.0179 | 1.0961 | -0.0034 | 0.0122 | 0.7471 | ||
75 | 75 | 0.0000 | 0.0068 | 0.5100 | -0.0001 | 0.0151 | 1.1325 | -0.0036 | 0.0101 | 0.7575 | ||
100 | 75 | 0.0002 | 0.0064 | 0.5543 | -0.0001 | 0.0136 | 1.1778 | -0.0039 | 0.0091 | 0.7881 | ||
50 | 100 | 0.0001 | 0.0070 | 0.4950 | 0.0009 | 0.0164 | 1.1597 | -0.0027 | 0.0103 | 0.7283 | ||
75 | 100 | 0.0000 | 0.0059 | 0.5110 | 0.0006 | 0.0130 | 1.1258 | -0.0025 | 0.0085 | 0.7361 | ||
100 | 100 | 0.0001 | 0.0050 | 0.5000 | -0.0002 | 0.0113 | 1.1300 | -0.0023 | 0.0075 | 0.7500 | ||
偏倚校正之后的表现 | ||||||||||||
Bias | RMSE | SRMSE | Bias | RMSE | SRMSE | Bias | RMSE | SRMSE | ||||
50 | 50 | -0.0001 | 0.0158 | 0.7900 | -0.0003 | 0.0088 | 0.4400 | 0.0007 | 0.0131 | 0.6550 | ||
75 | 50 | -0.0001 | 0.0126 | 0.7716 | 0.0002 | 0.0072 | 0.4409 | -0.0001 | 0.0105 | 0.6430 | ||
100 | 50 | -0.0004 | 0.0109 | 0.7707 | 0.0001 | 0.0063 | 0.4455 | 0.0000 | 0.0089 | 0.6293 | ||
50 | 75 | -0.0008 | 0.0124 | 0.7593 | 0.0003 | 0.0072 | 0.4409 | 0.0003 | 0.0107 | 0.6552 | ||
75 | 75 | 0.0001 | 0.0102 | 0.7650 | -0.0002 | 0.0061 | 0.4575 | -0.0002 | 0.0087 | 0.6525 | ||
100 | 75 | 0.0002 | 0.0095 | 0.8227 | -0.0001 | 0.0052 | 0.4503 | 0.0000 | 0.0076 | 0.6582 | ||
50 | 100 | -0.0007 | 0.0114 | 0.8061 | -0.0003 | 0.0062 | 0.4384 | 0.0000 | 0.0090 | 0.6364 | ||
75 | 100 | -0.0002 | 0.0088 | 0.7621 | -0.0002 | 0.0051 | 0.4417 | 0.0001 | 0.0074 | 0.6409 | ||
100 | 100 | 0.0000 | 0.0077 | 0.7700 | 0.0001 | 0.0043 | 0.4300 | -0.0002 | 0.0066 | 0.6600 | ||
Bias | RMSE | SRMSE | Bias | RMSE | SRMSE | Bias | RMSE | SRMSE | ||||
50 | 50 | -0.0012 | 0.0109 | 0.5450 | -0.0005 | 0.0229 | 1.1450 | -0.0003 | 0.0146 | 0.7300 | ||
75 | 50 | -0.0005 | 0.0085 | 0.5205 | -0.0003 | 0.0198 | 1.2125 | -0.0003 | 0.0120 | 0.7348 | ||
100 | 50 | 0.0006 | 0.0074 | 0.5233 | -0.0003 | 0.0160 | 1.1314 | -0.0004 | 0.0104 | 0.7354 | ||
50 | 75 | -0.0008 | 0.0084 | 0.5144 | 0.0009 | 0.0179 | 1.0961 | 0.0000 | 0.0119 | 0.7287 | ||
75 | 75 | 0.0000 | 0.0068 | 0.5100 | -0.0002 | 0.0151 | 1.1325 | -0.0003 | 0.0096 | 0.7200 | ||
100 | 75 | 0.0003 | 0.0064 | 0.5543 | -0.0002 | 0.0136 | 1.1778 | -0.0006 | 0.0084 | 0.7275 | ||
50 | 100 | 0.0001 | 0.0070 | 0.4950 | 0.0008 | 0.0164 | 1.1597 | -0.0002 | 0.0101 | 0.7142 | ||
75 | 100 | 0.0000 | 0.0059 | 0.5110 | 0.0005 | 0.0130 | 1.1258 | 0.0000 | 0.0082 | 0.7101 | ||
100 | 100 | 0.0001 | 0.0050 | 0.5000 | -0.0003 | 0.0112 | 1.1200 | 0.0002 | 0.0072 | 0.7200 |
表2 5%名义显著性水平下的实际(经验)显著性水平 |
N | T | λ | γ | β | ρ | σ2 | |
偏倚校之前的表现 | |||||||
50 | 50 | 6.7% | 10.3% | 5.5% | 6.1% | 5.5% | 10.2% |
75 | 50 | 5.1% | 11.4% | 4.9% | 4.5% | 7.2% | 11.3% |
100 | 50 | 4.9% | 14.0% | 5.6% | 5.5% | 5.8% | 11.6% |
50 | 75 | 4.2% | 6.3% | 5.9% | 4.5% | 4.3% | 8.9% |
75 | 75 | 5.3% | 10.9% | 6.6% | 5.4% | 5.4% | 8.9% |
100 | 75 | 7.4% | 12.1% | 6.4% | 7.5% | 6.0% | 9.8% |
50 | 100 | 6.4% | 7.2% | 5.1% | 4.2% | 6.2% | 7.6% |
5 | 100 | 3.9% | 8.9% | 4.3% | 4.6% | 4.9% | 7.4% |
100 | 100 | 5.1% | 9.5% | 7.0% | 4.1% | 5.3% | 7.6% |
偏倚校之后的表现 | |||||||
50 | 50 | 6.7% | 5.6% | 5.6% | 6.2% | 5.4% | 7.0% |
75 | 50 | 5.3% | 4.9% | 4.7% | 4.4% | 7.2% | 6.1% |
100 | 50 | 5.3% | 4.6% | 4.9% | 5.5% | 5.8% | 6.8% |
50 | 75 | 4.4% | 4.6% | 6.0% | 4.6% | 4.4% | 6.3% |
75 | 75 | 5.3% | 5.9% | 5.8% | 5.4% | 5.4% | 6.9% |
100 | 75 | 7.3% | 5.8% | 7.2% | 7.4% | 5.8% | 6.1% |
50 | 100 | 6.4% | 4.9% | 5.0% | 4.2% | 6.3% | 5.3% |
75 | 100 | 3.7% | 5.5% | 4.7% | 4.5% | 4.9% | 5.3% |
100 | 100 | 5.2% | 5.4% | 6.4% | 4.0% | 5.3% | 7.6% |
表3 脉冲响应函数95%名义显著性水平下的实际(经验)显著性水平 |
N | T | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |
单位扰动项改变导致的冲击响应表现 | |||||||||||
50 | 50 | 0.943 | 0.946 | 0.944 | 0.941 | 0.935 | 0.937 | 0.937 | 0.932 | 0.928 | 0.922 |
75 | 50 | 0.928 | 0.942 | 0.943 | 0.942 | 0.938 | 0.938 | 0.938 | 0.936 | 0.934 | 0.926 |
100 | 50 | 0.934 | 0.945 | 0.951 | 0.952 | 0.956 | 0.955 | 0.947 | 0.945 | 0.943 | 0.940 |
50 | 75 | 0.959 | 0.945 | 0.936 | 0.938 | 0.936 | 0.936 | 0.940 | 0.939 | 0.934 | 0.929 |
75 | 75 | 0.952 | 0.942 | 0.944 | 0.943 | 0.940 | 0.941 | 0.946 | 0.944 | 0.940 | 0.935 |
100 | 75 | 0.949 | 0.931 | 0.932 | 0.933 | 0.932 | 0.931 | 0.930 | 0.928 | 0.929 | 0.925 |
50 | 100 | 0.946 | 0.945 | 0.948 | 0.948 | 0.940 | 0.939 | 0.932 | 0.934 | 0.927 | 0.923 |
75 | 100 | 0.958 | 0.950 | 0.951 | 0.948 | 0.950 | 0.946 | 0.945 | 0.940 | 0.938 | 0.936 |
100 | 100 | 0.952 | 0.947 | 0.949 | 0.948 | 0.945 | 0.947 | 0.947 | 0.947 | 0.947 | 0.948 |
单位自变量改变导致的冲击响应表现 | |||||||||||
50 | 50 | 0.937 | 0.941 | 0.944 | 0.945 | 0.941 | 0.930 | 0.934 | 0.930 | 0.927 | 0.921 |
75 | 50 | 0.952 | 0.953 | 0.949 | 0.947 | 0.943 | 0.938 | 0.932 | 0.935 | 0.934 | 0.926 |
100 | 50 | 0.946 | 0.961 | 0.953 | 0.954 | 0.955 | 0.949 | 0.949 | 0.943 | 0.940 | 0.938 |
50 | 75 | 0.958 | 0.940 | 0.943 | 0.941 | 0.938 | 0.935 | 0.939 | 0.938 | 0.933 | 0.929 |
75 | 75 | 0.954 | 0.955 | 0.951 | 0.946 | 0.946 | 0.949 | 0.947 | 0.945 | 0.943 | 0.936 |
100 | 75 | 0.927 | 0.929 | 0.927 | 0.931 | 0.932 | 0.932 | 0.930 | 0.934 | 0.933 | 0.933 |
50 | 100 | 0.943 | 0.952 | 0.949 | 0.943 | 0.941 | 0.940 | 0.936 | 0.934 | 0.930 | 0.924 |
75 | 100 | 0.963 | 0.953 | 0.949 | 0.951 | 0.947 | 0.944 | 0.943 | 0.941 | 0.935 | 0.935 |
100 | 100 | 0.945 | 0.950 | 0.951 | 0.950 | 0.952 | 0.946 | 0.945 | 0.946 | 0.948 | 0.947 |
表4 区域收入差距的Moran's I指数 |
年 | Moran's I | z值 | p值 | 年 | Moran's I | z值 | p值 |
1996 | 0.377 | 3.472 | 0.00 | 2006 | 0.419 | 3.833 | 0.00 |
1997 | 0.369 | 3.406 | 0.00 | 2007 | 0.416 | 3.803 | 0.00 |
1998 | 0.362 | 3.349 | 0.00 | 2008 | 0.421 | 3.843 | 0.00 |
1999 | 0.360 | 3.329 | 0.00 | 2009 | 0.404 | 3.698 | 0.00 |
2000 | 0.374 | 3.448 | 0.00 | 2010 | 0.437 | 3.978 | 0.00 |
2001 | 0.373 | 3.441 | 0.00 | 2011 | 0.435 | 3.967 | 0.00 |
2002 | 0.381 | 3.508 | 0.00 | 2012 | 0.423 | 3.867 | 0.00 |
2003 | 0.402 | 3.684 | 0.00 | 2013 | 0.412 | 3.773 | 0.00 |
2004 | 0.406 | 3.721 | 0.00 | 2014 | 0.397 | 3.643 | 0.00 |
2005 | 0.417 | 3.812 | 0.00 | - | - | - | - |
表5 人力资本对区域经济差距的估计结果 |
变量 | 系数 | 变量 | 系数 | ||
校正前 | 校正后 | 校正前 | 校正后 | ||
W*GDP | 0.9012*** (42.33) | 0.9081*** (42.66) | Open | 0.0227* (1.74) | 0.0517*** (3.97) |
GDP-1 | 1.049*** (112.61) | 0.9672*** (103.79) | FDI | -0.0106 (-0.46) | 0.0305 (1.33) |
W*GDP-1 | -0.9712*** (-36.64) | -0.9386*** (-35.41) | Urban | 0.0865 (1.35) | -0.1983*** (-3.09) |
HCapital | 0.0124*** (3.08) | 0.0307*** (9.10) | -1.027*** (-13.12) | -1.0396*** (-13.28) | |
Inv | 0.0057** (2.01) | 0.0167*** (4.97) | 0.0050*** (3.46) | 0.0052*** (3.59) |
注: 括号内是 |
表6 累积效应 |
变量 | 直接效应 | 间接效应 | 总效应 |
HCapital | 1.1314*** (4.91) | -0.6465 (-0.64) | 0.4849 (0.47) |
Inv | 0.6164*** (9.43) | -0.3522 (-0.64) | 0.2642 (0.47) |
Open | 1.9049*** (3.83) | -1.0885 (-0.66) | 0.8164 (0.46) |
Urban | -7.3009*** (-3.30) | 4.1720 (0.58) | -3.1289 (-0.50) |
注: 括号内是 |
蔡昉, 都阳, 中国地区经济增长的趋同与差异——对西部开发战略的启示[J]. 经济研究, 2000, (10): 30- 37.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
邓飞, 柯文进, 异质型人力资本与经济发展——基于空间异质性的实证研究[J]. 统计研究, 2020, (2): 93- 104.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
范子英, 张军, 财政分权与中国经济增长的效率-基于非期望产出模型的分析[J]. 管理世界, 2009, (7): 15- 25.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
高远东, 花拥军, 人力资本空间效应与区域经济增长[J]. 地理研究, 2012, (4): 711- 719.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
龚六堂, 谢丹阳, 我国省份之间的要素流动和边际生产率的差异分析[J]. 经济研究, 2004, (1): 45- 53.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
李红权, 洪永淼, 汪寿阳, 我国A股市场与美股、港股的互动关系研究: 基于信息溢出视角[J]. 经济研究, 2011, (8): 15- 25.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
林毅夫, 刘明兴, 中国的经济增长收敛与收入分配[J]. 世界经济, 2003, (8): 3- 14.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
林毅夫, 刘培林, 中国的经济发展战略与地区收入差距[J]. 经济研究, 2003, (3): 19- 25.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
陆铭, 陈钊, 分割市场的经济增长——为什么经济开放可能加剧地方保护?[J]. 经济研究, 2009, (3): 42- 52.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
吕忠伟, R&D空间溢出对区域知识生产的作用研究[J]. 统计研究, 2009, (4): 44- 52.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
潘文卿, 中国的区域关联与经济增长的空间溢出效应[J]. 经济研究, 2012, (1): 54- 65.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
王小鲁, 樊纲, 中国地区差距的变动趋势和影响因素[J]. 经济研究, 2004, (1): 33- 44.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
肖志勇, 人力资本、空间溢出与经济增长——基于空间面板数据模型的经验分析[J]. 财经科学, 2010, (3): 61- 68.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
徐秋艳, 房胜飞, 马琳琳, 新型城镇化、产业结构升级与中国经济增长——基于空间溢出及门槛效应的实证研究[J]. 系统工程理论与实践, 2019, 39 (6): 1407- 1418.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
朱国忠, 乔坤元, 虞吉海, 中国各省经济增长是否收敛?[J]. 经济学(季刊), 2014, (3): 1171- 1194.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
LeSage J, Pace R, (2009). Introduction to Spatial Econometrics[M]. Chapman & Hall/CRC.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
Li K, (2018). Spatial Panel Data Models with Structural Change[R]. MPRA, No. 85388.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_ref.label}} |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
本文是李欣先博上学位论文部分章节的删减版,该文曾经在“首届中国计量经济学者论坛”上报告.
感谢李奇(Texas A&M)、廖明球、刘霞辉、田新民、王文举以及首经贸国际经管学院部分教师对初稿提出的修改建议.
/
〈 |
|
〉 |