So sánh kết quả với các nghiên cứu trên thế giới: Melo và cộng sự (2013) tổng
hợp 563 nghiên cứu, chủ yếu tập trung tại Mỹ và Châu Âu. Trong 563 nghiên cứu chỉ
có 26 nghiên cứu về đường hàng không, kết quả hệ số co giãn trung bình là 0,027.
Nghiên cứu này tại Việt Nam chưa tìm thấy ý nghĩa thống kê của đầu tư hàng không,
nguyên nhân vì hạ tầng giao thông hàng không của Việt Nam còn chưa phát triển, chỉ
một số ít tỉnh xây dựng sân bay, trong đó một một vài tỉnh có sân bay quốc tế. Theo
trên, trong 563 nghiên cứu có 27 nghiên cứu về cầu cảng, 32 nghiên cứu về đường sắt
và 282 nghiên cứu về đường bộ, giá trị hệ số co giãn trung bình của biến đại diện cho
đường bộ là 0,088, đường sắt là 0,037, cầu cảng là 0,068 và đường hàng không là
0,027. Do đó các hệ số tác động trực tiếp trong mô hình nghiên cứu không quá khác
biệt bên cạnh các khác biệt về biến đo lường đại diện
138 trang |
Chia sẻ: tueminh09 | Ngày: 08/02/2022 | Lượt xem: 450 | Lượt tải: 2
Bạn đang xem trước 20 trang tài liệu Luận án Các mô hình kinh tế lượng đánh giá tác động của hạ tầng giao thông vận tải đến phát triển kinh tế tại Việt Nam, để xem tài liệu hoàn chỉnh bạn click vào nút DOWNLOAD ở trên
ông gian và nhiều
dạng mô hình không gian.
- Kết hợp thêm các biến giải thích, thêm các biến trễ thời gian hoặc đánh giá
cho biến phụ thuộc là năng suất nhân tố tổng hợp (Total Factor Productivity - TFP).
101
DANH MỤC CÁC CÔNG TRÌNH TÁC GIẢ ĐÃ CÔNG BỐ
1. Lê Thị Quỳnh Nhung (2020), “Thực trạng vận tải hàng hóa đường biển khu vực
Bắc Trung Bộ và kinh nghiệm chính sách vận tải biển tại Nhật Bản”, Hội thảo
khoa học quốc gia Thực trạng phát triển và chính sách phát triển bền vững dải
ven biển vùng Bắc Trung Bộ, Đại học Kinh tế quốc dân, tháng 6/2020.
2. Lê Thị Quỳnh Nhung (2020), "Tác động lan tỏa không gian của vốn đầu tư giao
thông vận tải đến tăng trưởng kinh tế khu vực miền Trung", Tạp chí Khoa học và
Đào tạo Ngân hàng, Số 223, tháng 12/2020, trang 26 - 33.
3. Lê Thị Quỳnh Nhung (2021), “Tác động của các loại hình giao thông tới tăng trưởng
kinh tế: Nghiên cứu tại khu vực miền Nam”, Tạp chí Kinh tế và Dự báo, Số 9, tháng
3/2021, trang 128 - 131.
4. Lê Thị Quỳnh Nhung (2020), “Mô hình kinh tế lượng không gian đánh giá tác
động của vốn đầu tư giao thông đến tăng trưởng kinh tế khu vực đồng bằng sông
Hồng”, Tạp chí Công thương, Số 11/2020, trang 104 - 109.
5. Lê Thị Quỳnh Nhung (2020), “Tác động lan tỏa không gian của vốn đầu tư giao
thông đến kinh tế tại Trung du miền núi phía Bắc và so sánh với vùng đồng bằng
sông Hồng”, Tạp chí Công thương, Số 13/2020, trang 72 - 77.
102
TÀI LIỆU THAM KHẢO
1. Acemoglu, D. (2009), Introduction to modern economic growth, Princeton NJ:
Princeton University Press.
2. Acemoglu, D., and Johnson, S. (2005), 'Unbundling Institutions', Journal of
Political Economy, 113(5), 949 - 995.
3. Adams, S. (2009), 'Foreign Direct investment, domestic investment, and economic
growth in Sub-Saharan Africa', Journal of Policy Modeling, 31(6), 939 - 949.
4. Aisen, A., and Veiga, F. (2013), 'How does political instability affect economic
growth? European Journal of Political Economy, 29(C), 151 - 167.
5. Alataş, S., and Çakir, M. (2016), 'The Effect of Human Capital on Economic
Growth: A Panel Data Analysis', Journal of Administrative Sciences, 14, 539 - 555.
6. Anselin, L. (1988a), Spatial Econometrics: Methods and Models, Netherlands
and USA: Kluwer Academic Publicsers.
7. Anselin, L. (1988b), Spatial Econometrics: Methods and Models. Dordrecht, The
Netherlands: Kluwer Academic Publishers, Dordrecht, The Netherlands: Kluwer
Academic Publishers.
8. Anselin, L., Le Gallo, J., and Jayet, H. (2008), 'Spatial Panel Econometrics', The
Econometrics of Panel Data: Fundamentals and Recent Developments in Theory
and Practice (pp. 625 - 660), Berlin Heidelberg: Springer.
9. Arrow, K. J. a. K., M. . (1970), Public Investment, the Rate of Return, and
Optimal Fiscal Policy, Baltimore - USA: The Johns Hopkins University Press.
10. Arusha, C. V. (2009), 'Government Expenditure, Governance and Economic
Growth', Comparative Economic Studies, 51(3), 401 - 418.
11. Aschauer, D. A. (1989), 'Is Public Expenditure Productive?', Journal of Monetary
Economics, 23(2), 177 - 200.
12. Aschauer, D. A. (1990), Public investment and private sector growth: the
economic benefits of reducing America's "Third deficit", Washington DC.
13. Ayres, C. E. (1962), The theory of economic progress. A study of the Fundamental
Economic Development and Cultural Change, New York: Schocken.
103
14. Baltagi, B., and Pinnoi, N., (1995), 'Public Capital Stock and State Productivity
Growth: Further Evidence from an Error Components Model', Empirical
Economics, 20, 351 - 359.
15. Baltagi, B., P. Egger and M. Pfaffermayr. (2007), A Generalized Spatial Panel
Data Model with Random Effects, Syracuse University.
16. Barro, R. J. (1991), 'Economic growth in a cross section of countries', The
Quarterly Journal of Economics, 106(2), 407-443.
17. Barro, R. J. a. S.-i.-M., X. (1992), 'Convergence', Journal of Political Economy,
100(2), 223-251.
18. Becker, G. S. (1962), 'Investment in Human Capital: A Theoretical Analysis',
The Journal of Political Economy, 9-49, 70(5), 9 - 49.
19. Boarnet, M. G. (1996), 'The Direct and Indirect Economic Effects of Transportation
Infrastructure', Working Paper (Vol. 3, pp. 28), California, USA: Institute for
Transportation Studies - University of California.
20. Boopen, S. (2006), 'Transport Infrastructure and Economic Growth: Evidence
from Africa Using Dynamic Panel Estimates', The Empirical Economics Letters,
5(1), 37 - 52.
21. Bùi Thị Hoàng Lan. (2012), Mô hình nghiên cứu tác động của mạng lưới giao
thông đường bộ đến phát triển kinh tế- xã hội của vùng kinh tế trọng điểm Bắc
Bộ, Luận án tiến sĩ, Đại học Kinh tế quốc dân.
22. Canning, D. (1999), 'Infrastructure’s contribution to aggregate output', World
Bank Policy Research, 2246.
23. Chí Tín, Thu Phương (2018), Bàn giải pháp nâng tầm du lịch Việt qua kết nối
nhiều điểm đến, nguồn bản đồ Việt Nam truy cập ngày 10/8/2019 tại
https://baodautu.vn/ban-giai-phap-nang-tam-du-lich-viet-qua-ket-noi-nhieu-
diem-den-d85830.html
24. Cliff, A. D., and Ord, J. K. (1981), Spatial Processes, Models and Applications,
London: Pion Ltd.
25. Cobb, C. W., and Douglas, Paul H. (1928), 'A theory of production', American
Economic Review, 18(1), 1939 - 1965.
104
26. Đào Thông Minh, Lê Thị Mai Hương. (2016), 'Nghiên cứu tác động của vốn đầu
tư tư nhân, lao động và cơ sở hạ tầng đến tăng trưởng kinh tế vùng Đồng bằng
Sông Cửu Long', Van Hien University Journal of Science, 4(3), 65 - 74.
27. De la Fuente, A., and Doménech, R. (2006), 'Human capital in growth
regressions: how much difference does data quality make?', Journal of the
European Economic Association, 4(1), 1 - 36.
28. Demurger, S. (2001), 'Infrastructure development and economic growth: an
explanation for regional disparities in China?', Journal of Comparative
Economics, 29(1), 95-117.
29. Deng, T., Shao, S., Yang, L. and Zhang, X. (2014), 'Has the transport-led
economic growth effect reached a peak in China? A panel threshold regression
approach', Transportation, 41(3), 567 - 587.
30. Dương Bích Tiến. (1998), Nghiên cứu đánh giá thực trạng của hệ thống hạ tầng
cơ sở giao thông và phân tích ảnh hưởng của nó đến các nghành khác, Đề tài cấp
Ngành, Vụ Xây dựng cơ bản - Giao thông - Bưu điện - Tổng cục Thống kê
31. Eisner, R. (1991), 'Infrastructure and regional economic performance: comment',
New England Economic Review, September/October, 47-58.
32. Filip, N., and Popa, C. (2014), 'The role of transportation in economic
development', “Mircea cel Batran” Naval Academy Scientific Bulletin, XVII(2).
33. Fingleton, B. (2008), 'A Generalized Method of Moments Estimators for a
Spatial Panel Model with an Endogenous Spatial Lag and Spatial Moving
Average Errors', Spatial Economic Analysis, 3, 27-44.
34. Finn, M. (1993), 'Is all government capital productive?', Economic Quarterly,
79(4), 53-80.
35. Grossman, H. I. a. L., R. (1974), 'The Macro-Economic Effects of Productive
Public Expenditures', The Manchester School of Economics and Social Studies,
46(2), 162-170.
36. Hansen, L. (1982), 'Large Sample Properties of Generalized Method of Moment
Estimators', Econometrica, 50, 1029-1054.
37. Harrod, R. F. (1937), 'Mr. Keynes and Traditional Theory', Econometrica, 5(1),
74 - 86.
38. Harrod, R. F. (1973), Economic Dynamics, London: Macmillan.
105
39. Hicks, J. (1932), The Theory of Wages, Macmillan, London.
40. Holtz-Eakin, D., and Schwartz, A. (1995), 'Spatial Productivity Spillovers from
Public Infrastructure: Evidence from State Highways', International Tax and
Public Finance, 2(3), 459 - 468.
41. Hong, J., Chu, J. and Wang, Q. (2011), 'Transport infrastructure and regional
economic growth: evidence from China', Transportation, 38(5), 737 - 752.
42. Hu, Z. and Lou, S. (2017), 'Road infrastructure, spatial spillover and county economic
growth', IOP Conference Series: Materials Science and Engineering, 231, 7p.
43. Jiang, X., Zhang, L., Xiong, C., and Wang, R. (2015), 'Transportation and
Regional Economic Development: Analysis of Spatial Spillovers in China
Provincial Regions', Networks and Spatial Economics, 16(3), 769 - 790.
44. Jiwattanakulpaisarn, P., Noland, R. B., & Graham, D. J. (2011), 'Highway
infrastructure and private output: Evidence from static and dynamic production
functionmodels', Transportmetrica, 7(5), 347 - 367.
45. Kapoor, M., Kelejian, H.H. and I.R. Prucha. (2007), 'Panel Data Models with
Spatially Correlated Error Components', Journal of Econometrics, 140, 97-130.
46. Lensink, R., Hong Bo, and Elmer Sterken. (1999), 'Does Uncertainty Affect
Economic Growth? An Empirical Analysis', Weltwirtschaftliches Archiv, 135,
379 - 396.
47. Lewis, A. (1955), The Theory of Economic Growth, London: George Allen &
Unwin Ltd.
48. Li, J., Wen, J. and Jiang, B. (2017), 'Spatial Spillover Effects of Transport
Infrastructure in Chinese New Silk Road Economic Belt', International Journal
of e-Navigation and Maritime Economy, 6, 1 - 8.
49. Ligthart, J. (2000), 'Public capital and output growth in Portugal: an empirical
analysis', IMF Working Paper, 11, 1-38.
50. MaCurdy, T., and Pencavel, J. (1986), 'Testing between Competing Models of
Wage and Employment Determination in Unionized Markets', Journal of
Political Economy, 94(2, Part 2), S3 - S39.
51. Melo, P. C., and Graham, D., and Ardao, R. (2013), 'The Productivity of
Transport Infrastructure Investment: A Meta-Analysis of Empirical Evidence',
Regional Science and Urban Economics, 43(5), 695 - 706.
106
52. Moran, P. (1950), 'Notes on Continuous Stochastic Phenomena', Biometrika, 37, 17-23.
53. Munnell, A. H. (1990a), 'Why Has Productivity Growth Declined? Productivity
and Public Investment', New England Economic Review, Jan./Feb., 3-20.
54. Munnell, A. H., and Cook, L. M. (1990b), 'How Does Public Infrastructure
Affect Regional Economic Performance', New England Economic Review,
Sept./Oct., 11 - 33.
55. Munnell, A. H. a. C., L. M. (1990b), 'How Does Public Infrastructure Affect Regional
Economic Performance', New England Economic Review, Sept./Oct., 11-33.
56. Murphy, K. M., Andrei Shleifer, and Robert W. Vishny. (1993), 'Why is RentSeeking
So Costly to Growth?', American Economic Review, 84(2), 409 - 414.
57. Ngô Anh Tín. (2017), Tác động đầu tư công đến tăng trưởng kinh tế vùng đồng
bằng sông Cửu Long, Luận án tiến sĩ, Trường Đại học Kinh tế TP. Hồ Chí Minh,
TP Hồ Chí Minh.
58. Nguyễn Phan Thu Hằng. (2016), 'Vai trò nguồn nhân lực chất lượng cao trong
thúc đẩy sáng tạo và ứng dụng khoa học - công nghệ', Science & Technology
Development, 19, 30 - 40.
59. Nguyễn Văn Tạo. (2009), 'Việt Nam trước khủng hoảng và suy thoái kinh tế toàn
cầu', Tạp chí nghiên cứu Kinh tế và Kinh doanh Châu Á (Tạp chí Phát triển kinh
tế), 220 (Tháng 2), 125 - 132.
60. Pande, R. (2009), 'Understanding Political Corruption in Low Income Countries',
Chapter 50 in Handbook of Development Economics (Vol. 4, pp. 3155-3184),
Amsterdam, North Holland, Elsevier.
61. Petty, W. (1900), 'Petty's Place in the History of Economic Theory', The
Quarterly Journal of Economics, 14(3), 307-340.
62. Phan Trang (2020), Trục cao tốc xuyên Việt đầu tiên đang dần hình thành, tại
thanh/408933.vgp
63. Phùng Trọng (2016), Đẩy nhanh tiến độ Dự án đầu tư xây dựng nâng cấp tuyến
kênh Chợ Gạo, nguồn ảnh kênh chợ Gạo truy cập ngày 10/8/2020 tại
dung-nang-cap-tuyen-kenh-cho-gao.aspx - Nguồn ảnh kênh chợ Gạo.
107
64. Ramsey, F. (1928), 'A Mathematical Theory of Saving', Economic Journal, 38,
543-559.
65. Ricardo, D. (1951 - 1973), The Works and Correspondence of David Ricardo, 11
Vols. Cambridge: Cambridge University Press.
66. Robinson, J. (1938), 'The Classification of Inventions', Review of Economic
Studies, 5(2), 139 - 142.
67. Rodrik, D., Subramanian, A. and Trebbi, F. (2004), 'Institutions Rule: The
Primacy of Institutions Over Geography and Integration in Economic
Development', Journal of Economic Growth, 9, 131 - 165.
68. Romer, R. (1989), 'Human Capital and Growth: Theory and Evidence', NBER
Working Paper Series, 3173, 1 - 41.
69. Rosenberg, N. (1968). Adam Smith, consumer tastes, and economic growth.
Journal of Political Economy, 76(3), 361 - 374.
70. SACTRA (Standing Advisory Committee on Trunk Road Assessment), UK
Department of the Environment, Transport and the Regions. (1999), Transport
and the Economy, London.
71. Schultz, T. W. (1961), 'Investment in Human Capital', American Economic
Review, 51(1), 1 - 17.
72. Serven, L. (1996), 'Does Public Capital Crowd Out Private Capital? Evidence
from India', Policy Research Working Paper, 1613.
73. Silva, A. R. d. (2018), Creating a Spatial Weights Matrix for the SPATIALREG
Procedure, Paper presented at the SAS Global Forum 2018, Denver, Colorado.
74. Smith, A. (1976), An Inquiry into the Nature and Causes of the Wealth of
Nations, Oxford: Oxford University Press.
75. Smith, A. (1998), The Wealth of Nations, Oxford University Press.
76. Solow, R. (1969), Investment and technical change, Mathematical Models in the
Social Sciences, Stanford University Press, Palo Alto.
77. Solow, R. M. (1956), 'A Contribution to the Theory of Economic Growth',
Quarterly Journal of Economics, 70, 65-94.
108
78. Sushil Kumar Haldar and Girijasankar Mallik. (2010), 'Does Human Capital
Cause Economic Growth? A Case of India', International Journal of Economic
Sciences and Applied Research, 3(1), 7 - 25.
79. Svensson, J. (2003), 'Who Must Pay Bribes and How Much?', Quarterly Journal
of Economics, 118(1), 207-230.
80. Swan, T. W. (1956), 'Economic Growth and Capital Accumulation', Economic
Record, 32, 334-361.
81. Thompson, E. A. (1972), 'Review: Public Investment, The Rate of Return, and
Optimal Fiscal Policy by Kenneth J. Arrow; Mordecai Kurz', Economic Society,
40(6), 1174-1176.
82. Tổng cục Thống kê. (2013), Tình hình kinh tế - xã hội năm 2012, tại
ategoryId=100002607&articleId=10051323
83. Turkey, J. (1957), 'On the comparative anatomy of transformations', The Annals
of Mathematical Statistics, 28(3), 602 - 632.
84. Uzawa, H. (1961), 'Neutral Inventions and the Stability of Growth Equilibrium',
Review of Economic Studies, 28(2), 117 - 124.
85. Vega, H., and Elhorst, J.P. (2015), 'The SLX model', Journal of Regional
Science, 55, 339 - 363
86. Vũ Hoàng Dương, Phí Vĩnh Tường và Phạm Sỹ An. (2014), 'Mối quan hệ giữa đầu tư
phát triển hạ tầng giao thông và tăng trưởng', Nghiên cứu Kinh tế, 433, 20 - 29.
87. Wicksell, K. (1916), 'Den 'kritiska punkten' i lagen for jordbrukets aftagande
produktivitet, Economisk Tidskrift, 18(8), 285 - 292.
88. Wilson, A. (1970), Entropy in Urban and Regional Modelling, London: Pion Press.
89. Yamaguchi, K. (2007), 'Inter-regional air transport accessibility and macro-
economic performance in Japan', Transportation Research, 43(3), 247 - 258.
90. Zou, W., Zhang, F., Zhuang, Z., and Song,. H. (2008), 'Transport Infrastructure,
Growth, and Poverty Alleviation: Empirical Analysis of China', Annals of
Economics and Finance, 9(2), 345 - 371.
109
PHỤ LỤC
Mô hình 1a xét tác động của của vốn đầu tư cho giao thông
Kiểm định Lagrange (LM) lựa chọn giữa mô hình POLS và RE
rho .96012822 (fraction of variance due to u_i)
sigma_e .08173262
sigma_u .40107654
_cons -2.193325 2.35585 -0.93 0.355 -6.902602 2.515953
D .0790461 .0156644 5.05 0.000 .0477334 .1103588
LnTransparency .1443488 .0661093 2.18 0.033 .0121982 .2764994
LnEntryCosts .1629524 .0415514 3.92 0.000 .0798923 .2460125
LnLTraining_PCI .2610303 .0518631 5.03 0.000 .1573574 .3647032
LnTLRate .2281022 .0486433 4.69 0.000 .1308657 .3253387
LnL 1.068741 .1822488 5.86 0.000 .7044307 1.433051
LnKT .0478914 .0140482 3.41 0.001 .0198094 .0759734
LnKNT .1674246 .0336373 4.98 0.000 .1001845 .2346646
LnGRDP Coef. Std. Err. t P>|t| [95% Conf. Interval]
Robust
(Std. Err. adjusted for 63 clusters in ID)
corr(u_i, Xb) = 0.0959 Prob > F = 0.0000
F(8,62) = 111.44
overall = 0.8133 max = 8
between = 0.8120 avg = 8.0
within = 0.8439 min = 8
R-sq: Obs per group:
Group variable: ID Number of groups = 63
Fixed-effects (within) regression Number of obs = 504
> fe rob
. xtreg LnGRDP LnKNT LnKT LnL LnTLRate LnLTraining_PCI LnEntryCosts LnTransparency D,
Prob > chibar2 = 0.0000
chibar2(01) = 1319.91
Test: Var(u) = 0
u .1139232 .3375251
e .0066802 .0817326
LnGRDP .8725719 .9341156
Var sd = sqrt(Var)
Estimated results:
LnGRDP[ID,t] = Xb + u[ID] + e[ID,t]
Breusch and Pagan Lagrangian multiplier test for random effects
. xttest0
110
Kiểm định Hausman lựa chọn mô hình 1a
Kiểm định đa cộng tuyến
(V_b-V_B is not positive definite)
Prob>chi2 = 0.0001
= 29.48
chi2(7) = (b-B)'[(V_b-V_B)^(-1)](b-B)
Test: Ho: difference in coefficients not systematic
B = inconsistent under Ha, efficient under Ho; obtained from xtreg
b = consistent under Ho and Ha; obtained from xtreg
D .0790461 .0744354 .0046107 .0058568
LnTranspar~y .1443488 .1456669 -.0013181 .0049515
LnEntryCosts .1629524 .1583321 .0046204 .0077151
LnLTrainin~I .2610303 .2546865 .0063438 .002958
LnTLRate .2281022 .2374498 -.0093476 .009424
LnL 1.068741 1.055879 .0128623 .1119366
LnKT .0478914 .0474866 .0004048 .0018475
LnKNT .1674246 .1831566 -.0157321 .0034525
fe re Difference S.E.
(b) (B) (b-B) sqrt(diag(V_b-V_B))
Coefficients
coefficients are on a similar scale.
anything unexpected and possibly consider scaling your variables so that the
be problems computing the test. Examine the output of your estimators for
coefficients being tested (8); be sure this is what you expect, or there may
Note: the rank of the differenced variance matrix (7) does not equal the number of
. hausman fe re, sig
Mean VIF 2.42
----------------------------------------------------
D 1.40 1.18 0.7135 0.2865
LnTransparency 1.23 1.11 0.8120 0.1880
LnEntryCosts 1.14 1.07 0.8808 0.1192
LnLTraining_PCI 3.06 1.75 0.3271 0.6729
LnTLRate 2.97 1.72 0.3371 0.6629
LnL 3.33 1.83 0.2999 0.7001
LnKT 2.03 1.42 0.4936 0.5064
LnKNT 4.17 2.04 0.2396 0.7604
----------------------------------------------------
Variable VIF VIF Tolerance Squared
SQRT R-
Collinearity Diagnostics
(obs=504)
. collin LnKNT LnKT LnL LnTLRate LnLTraining_PCI LnEntryCosts LnTransparency D
111
Mô hình không gian SLX 1b xét tác động của vốn đầu tư cho giao thông
Wald test of spatial terms: chi2(1) = 34.44 Prob > chi2 = 0.0000
/sigma_e .0776008 .0026188 .0726342 .082907
/sigma_u .4107921 .037328 .3437754 .4908733
LnKT .1629157 .0277625 5.87 0.000 .1085022 .2173292
M
_cons -3.460151 1.006441 -3.44 0.001 -5.432739 -1.487563
D .0343479 .013518 2.54 0.011 .0078531 .0608426
LnTransparency .0941506 .0424176 2.22 0.026 .0110136 .1772876
LnEntryCosts .1351351 .0370573 3.65 0.000 .0625042 .207766
LnLTraining_PCI .1721441 .0470981 3.66 0.000 .0798336 .2644546
LnTLRate .1980595 .0342482 5.78 0.000 .1309344 .2651846
LnL 1.039943 .0770859 13.49 0.000 .8888576 1.191029
LnKT .0352967 .0090536 3.90 0.000 .0175519 .0530415
LnKNT .1659842 .0170036 9.76 0.000 .1326577 .1993107
LnGRDP
LnGRDP Coef. Std. Err. z P>|z| [95% Conf. Interval]
Log likelihood = 402.5360 Pseudo R2 = 0.8036
Prob > chi2 = 0.0000
Wald chi2(9) = 2872.36
Obs per group = 8
Group variable: ID Number of groups = 63
Random-effects spatial regression Number of obs = 504
Iteration 1: log likelihood = 402.53596 (backed up)
Iteration 0: log likelihood = 402.53596
Optimizing unconcentrated log likelihood:
Iteration 4: log likelihood = 402.53596
Iteration 3: log likelihood = 402.53596
Iteration 2: log likelihood = 402.53586
Iteration 1: log likelihood = 402.50387
Iteration 0: log likelihood = 398.13307
rescale eq: log likelihood = 398.13307
rescale: log likelihood = 398.13307
improve: log likelihood = 398.13307
initial: log likelihood = 398.13307
Optimizing concentrated log likelihood:
(weighting matrix defines 63 places)
(data contain 63 panels (places) )
(504 observations used)
(504 observations)
> cy D, ivarlag(M: LnKT) re
. spxtregress LnGRDP LnKNT LnKT LnL LnTLRate LnLTraining_PCI LnEntryCosts LnTransparen
112
Mô hình RE 2a xét tác động của vốn đầu tư cho giao thông
tại khu vực Miền Bắc
rho .87393198 (fraction of variance due to u_i)
sigma_e .08345873
sigma_u .21973949
_cons -4.576295 .7616116 -6.01 0.000 -6.069026 -3.083564
D .0717648 .02219 3.23 0.001 .0282732 .1152563
LnTransparency .2413178 .0883342 2.73 0.006 .0681859 .4144497
LnEntryCosts .1188436 .0547557 2.17 0.030 .0115244 .2261628
LnTLRate .4216475 .0851241 4.95 0.000 .2548073 .5884877
LnL 1.063045 .0831968 12.78 0.000 .8999822 1.226108
LnKT .0937781 .0128577 7.29 0.000 .0685775 .1189787
LnKNT .2620777 .0415305 6.31 0.000 .1806794 .343476
LnGRDP Coef. Std. Err. z P>|z| [95% Conf. Interval]
Robust
(Std. Err. adjusted for 25 clusters in ID)
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000
Wald chi2(7) = 1744.50
overall = 0.9226 max = 8
between = 0.9254 avg = 8.0
within = 0.8624 min = 8
R-sq: Obs per group:
Group variable: ID Number of groups = 25
Random-effects GLS regression Number of obs = 200
> b
. xtreg LnGRDP LnKNT LnKT LnL LnTLRate LnEntryCosts LnTransparency D if ID <=25, re ro
113
Mô hình không gian SLX 2b xét tác động của vốn đầu tư cho giao thông tại khu
vực Miền Bắc
Wald test of spatial terms: chi2(1) = 21.53 Prob > chi2 = 0.0000
/sigma_e .0785353 .0042244 .0706771 .0872673
/sigma_u .2882519 .0429361 .2152696 .3859772
LnKT .2285329 .0492549 4.64 0.000 .1319951 .3250707
M_s001
_cons -7.333448 1.337126 -5.48 0.000 -9.954167 -4.712728
D .010574 .0214951 0.49 0.623 -.0315557 .0527036
LnTransparency .1694829 .0583271 2.91 0.004 .0551639 .283802
LnEntryCosts .0739235 .0534979 1.38 0.167 -.0309305 .1787775
LnTLRate .3317214 .0631559 5.25 0.000 .2079381 .4555046
LnL 1.12549 .1010597 11.14 0.000 .9274165 1.323563
LnKT .080307 .015968 5.03 0.000 .0490103 .1116038
LnKNT .2259086 .0253508 8.91 0.000 .1762219 .2755953
LnGRDP
LnGRDP Coef. Std. Err. z P>|z| [95% Conf. Interval]
Log likelihood = 166.4380 Pseudo R2 = 0.9138
Prob > chi2 = 0.0000
Wald chi2(8) = 1550.66
Obs per group = 8
Group variable: ID Number of groups = 25
Random-effects spatial regression Number of obs = 200
Iteration 1: log likelihood = 166.43805 (backed up)
Iteration 0: log likelihood = 166.43805
Optimizing unconcentrated log likelihood:
Iteration 4: log likelihood = 166.43805
Iteration 3: log likelihood = 166.43805
Iteration 2: log likelihood = 166.43796
Iteration 1: log likelihood = 166.41626
Iteration 0: log likelihood = 163.87094
rescale eq: log likelihood = 163.87094
rescale: log likelihood = 163.64097
improve: log likelihood = 163.64097
initial: log likelihood = 163.64097
Optimizing concentrated log likelihood:
(weighting matrix M_s001 created)
(weighting matrix matched 25 places in data)
(you specified -force-)
(weighting matrix defines 63 places)
(data contain 25 panels (places) )
(200 observations used)
(304 observations excluded due to if/in)
(504 observations)
> varlag(M: LnKT) force re
. spxtregress LnGRDP LnKNT LnKT LnL LnTLRate LnEntryCosts LnTransparency D if ID <=25, i
114
Mô hình RE 3a xét tác động của vốn đầu tư cho giao thông
tại khu vực Miền Trung
rho .93813007 (fraction of variance due to u_i)
sigma_e .06876409
sigma_u .26776475
_cons 1.155199 1.434276 0.81 0.421 -1.655931 3.966328
D .0708677 .019105 3.71 0.000 .0334226 .1083127
LnEntryCosts .110939 .0561999 1.97 0.048 .0007892 .2210888
LnLTraining_PCI .2483464 .08666 2.87 0.004 .078496 .4181969
LnTLRate .3915474 .0807927 4.85 0.000 .2331966 .5498982
LnL .9183692 .1122948 8.18 0.000 .6982756 1.138463
LnKT .0407635 .0190889 2.14 0.033 .0033499 .0781771
LnKNT .0815693 .0270603 3.01 0.003 .0285321 .1346065
LnGRDP Coef. Std. Err. z P>|z| [95% Conf. Interval]
Robust
(Std. Err. adjusted for 19 clusters in ID)
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000
Wald chi2(7) = 760.52
overall = 0.8428 max = 8
between = 0.8395 avg = 8.0
within = 0.8846 min = 8
R-sq: Obs per group:
Group variable: ID Number of groups = 19
Random-effects GLS regression Number of obs = 152
> 44, re rob
. xtreg LnGRDP LnKNT LnKT LnL LnTLRate LnLTraining_PCI LnEntryCosts D if ID>=26& ID<=
115
Mô hình không gian SLX 3b xét tác động của vốn đầu tư cho giao thông
tại Miền Trung
.
Wald test of spatial terms: chi2(1) = 9.80 Prob > chi2 = 0.0017
/sigma_e .065354 .0040212 .0579293 .0737303
/sigma_u .2324908 .0390178 .1673219 .3230418
LnKT .0889112 .0284072 3.13 0.002 .0332342 .1445882
M_s001
_cons .8844324 1.227635 0.72 0.471 -1.521687 3.290552
D .0504895 .01936 2.61 0.009 .0125447 .0884344
LnEntryCosts .0908292 .0596399 1.52 0.128 -.0260629 .2077213
LnLTraining_PCI .1664713 .0786795 2.12 0.034 .0122624 .3206803
LnTLRate .3494226 .0568063 6.15 0.000 .2380842 .460761
LnL .8848519 .0948925 9.32 0.000 .6988659 1.070838
LnKT .0291794 .0123505 2.36 0.018 .0049729 .0533859
LnKNT .0797412 .0270514 2.95 0.003 .0267214 .132761
LnGRDP
LnGRDP Coef. Std. Err. z P>|z| [95% Conf. Interval]
Log likelihood = 155.0081 Pseudo R2 = 0.8388
Prob > chi2 = 0.0000
Wald chi2(8) = 1185.44
Obs per group = 8
Group variable: ID Number of groups = 19
Random-effects spatial regression Number of obs = 152
Iteration 1: log likelihood = 155.00808 (backed up)
Iteration 0: log likelihood = 155.00808
Optimizing unconcentrated log likelihood:
Iteration 3: log likelihood = 155.00808
Iteration 2: log likelihood = 155.00808
Iteration 1: log likelihood = 155.00802
Iteration 0: log likelihood = 154.96869
rescale eq: log likelihood = 154.96869
rescale: log likelihood = 154.96869
improve: log likelihood = 154.96869
initial: log likelihood = 154.96869
Optimizing concentrated log likelihood:
(weighting matrix M_s001 created)
(weighting matrix matched 19 places in data)
(you specified -force-)
(weighting matrix defines 63 places)
(data contain 19 panels (places) )
(152 observations used)
(352 observations excluded due to if/in)
(504 observations)
> D<= 44, ivarlag(M: LnKT) force re
. spxtregress LnGRDP LnKNT LnKT LnL LnTLRate LnLTraining_PCI LnEntryCosts D if ID>=26& I
116
Mô hình FE 4a xét tác động của vốn đầu tư cho giao thông tại khu vực Miền Nam
rho .98361262 (fraction of variance due to u_i)
sigma_e .07327428
sigma_u .56768679
_cons 4.259601 3.65163 1.17 0.259 -3.41219 11.93139
D .1185983 .0203848 5.82 0.000 .0757714 .1614253
LnEntryCosts .2848703 .0600305 4.75 0.000 .1587509 .4109898
LnLTraining_PCI .257399 .0938901 2.74 0.013 .0601432 .4546547
LnTLRate .1302805 .0542926 2.40 0.027 .0162161 .244345
LnL .6461228 .3024393 2.14 0.047 .0107214 1.281524
LnKT .0290284 .0203663 1.43 0.171 -.0137596 .0718165
LnKNT .1694531 .0655354 2.59 0.019 .0317684 .3071378
LnGRDP Coef. Std. Err. t P>|t| [95% Conf. Interval]
Robust
(Std. Err. adjusted for 19 clusters in ID)
corr(u_i, Xb) = 0.5707 Prob > F = 0.0000
F(7,18) = 62.25
overall = 0.7741 max = 8
between = 0.7820 avg = 8.0
within = 0.8421 min = 8
R-sq: Obs per group:
Group variable: ID Number of groups = 19
Fixed-effects (within) regression Number of obs = 152
> ob
. xtreg LnGRDP LnKNT LnKT LnL LnTLRate LnLTraining_PCI LnEntryCosts D if ID >=45, fe r
117
Mô hình không gian SLX 4b xét tác động của vốn đầu tư cho giao thông
tại Miền Nam
Wald test of spatial terms: chi2(1) = 11.53 Prob > chi2 = 0.0007
/sigma_e .0684263 .0042257 .0606256 .0772307
/sigma_u .5328169 .0908963 .3813876 .7443709
LnKT .1512496 .0445466 3.40 0.001 .0639399 .2385594
M_s001
_cons .9696892 2.083454 0.47 0.642 -3.113805 5.053184
D .0616837 .0233766 2.64 0.008 .0158664 .1075009
LnEntryCosts .230762 .0671657 3.44 0.001 .0991197 .3624043
LnLTraining_PCI .1326153 .0864264 1.53 0.125 -.0367774 .302008
LnTLRate .1413867 .0494352 2.86 0.004 .0444955 .238278
LnL .7783511 .1614414 4.82 0.000 .4619319 1.09477
LnKT .0290602 .0156394 1.86 0.063 -.0015925 .059713
LnKNT .1531595 .0300279 5.10 0.000 .0943058 .2120131
LnGRDP
LnGRDP Coef. Std. Err. z P>|z| [95% Conf. Interval]
Log likelihood = 133.2148 Pseudo R2 = 0.7604
Prob > chi2 = 0.0000
Wald chi2(8) = 811.09
Obs per group = 8
Group variable: ID Number of groups = 19
Random-effects spatial regression Number of obs = 152
Iteration 1: log likelihood = 133.21481 (backed up)
Iteration 0: log likelihood = 133.21481
Optimizing unconcentrated log likelihood:
Iteration 4: log likelihood = 133.21481
Iteration 3: log likelihood = 133.21481
Iteration 2: log likelihood = 133.21429
Iteration 1: log likelihood = 132.95051
Iteration 0: log likelihood = 132.74186
rescale eq: log likelihood = 132.74186
rescale: log likelihood = 126.8262
improve: log likelihood = 126.8262
initial: log likelihood = 126.8262
Optimizing concentrated log likelihood:
(weighting matrix M_s001 created)
(weighting matrix matched 19 places in data)
(you specified -force-)
(weighting matrix defines 63 places)
(data contain 19 panels (places) )
(152 observations used)
(352 observations excluded due to if/in)
(504 observations)
> ivarlag(M: LnKT) force re
. spxtregress LnGRDP LnKNT LnKT LnL LnTLRate LnLTraining_PCI LnEntryCosts D if ID >=45,
118
Mô hình FE 5a xét tác động của các loại hình vốn đầu tư cho giao thông
Kiểm định Lagrange (LM) lựa chọn giữa mô hình POLS và RE
rho .96017169 (fraction of variance due to u_i)
sigma_e .08232476
sigma_u .40421178
_cons -2.305118 2.311302 -1.00 0.322 -6.925344 2.315108
D .0784492 .0160189 4.90 0.000 .0464279 .1104705
LnTransparency .1378471 .0673922 2.05 0.045 .0031321 .2725622
LnEntryCosts .1500667 .041498 3.62 0.001 .0671133 .2330201
LnLTraining_PCI .2579381 .0509658 5.06 0.000 .1560589 .3598173
LnTLRate .2320662 .0501155 4.63 0.000 .1318869 .3322456
LnL 1.087174 .1794388 6.06 0.000 .728481 1.445867
LnKWTSupport .0070317 .0023791 2.96 0.004 .0022758 .0117875
LnKPRRoad .0324438 .0133951 2.42 0.018 .0056674 .0592203
LnKNT .1704863 .0335438 5.08 0.000 .1034332 .2375394
LnGRDP Coef. Std. Err. t P>|t| [95% Conf. Interval]
Robust
(Std. Err. adjusted for 63 clusters in ID)
corr(u_i, Xb) = 0.0632 Prob > F = 0.0000
F(9,62) = 100.92
overall = 0.8094 max = 8
between = 0.8080 avg = 8.0
within = 0.8420 min = 8
R-sq: Obs per group:
Group variable: ID Number of groups = 63
Fixed-effects (within) regression Number of obs = 504
> LnTransparency D, fe rob
. xtreg LnGRDP LnKNT LnKPRRoad LnKWTSupport LnL LnTLRate LnLTraining_PCI LnEntryCosts
Prob > chibar2 = 0.0000
chibar2(01) = 1267.36
Test: Var(u) = 0
u .1056404 .3250236
e .0067774 .0823248
LnGRDP .8725719 .9341156
Var sd = sqrt(Var)
Estimated results:
LnGRDP[ID,t] = Xb + u[ID] + e[ID,t]
Breusch and Pagan Lagrangian multiplier test for random effects
. xttest0
119
Kiểm định Hausman lựa chọn giữa mô hình FE và RE
Kiểm định đa cộng tuyến trong mô hình FE 5a
Prob>chi2 = 0.0000
= 36.96
chi2(8) = (b-B)'[(V_b-V_B)^(-1)](b-B)
Test: Ho: difference in coefficients not systematic
B = inconsistent under Ha, efficient under Ho; obtained from xtreg
b = consistent under Ho and Ha; obtained from xtreg
D .0784492 .0744451 .0040041 .0060888
LnTranspar~y .1378471 .1390289 -.0011818 .0051412
LnEntryCosts .1500667 .1433528 .0067139 .0080696
LnLTrainin~I .2579381 .2504425 .0074956 .0031973
LnTLRate .2320662 .2460342 -.0139679 .0100428
LnL 1.087174 1.058961 .0282128 .1147218
LnKWTSupport .0070317 .0085292 -.0014975 .0004112
LnKPRRoad .0324438 .027291 .0051528 .0018369
LnKNT .1704863 .1874631 -.0169768 .0036039
fe re Difference S.E.
(b) (B) (b-B) sqrt(diag(V_b-V_B))
Coefficients
coefficients are on a similar scale.
anything unexpected and possibly consider scaling your variables so that the
be problems computing the test. Examine the output of your estimators for
coefficients being tested (9); be sure this is what you expect, or there may
Note: the rank of the differenced variance matrix (8) does not equal the number of
. hausman fe re, sig
Mean VIF 2.37
----------------------------------------------------
D 1.40 1.18 0.7128 0.2872
LnTransparency 1.23 1.11 0.8115 0.1885
LnEntryCosts 1.15 1.07 0.8703 0.1297
LnLTraining_PCI 3.07 1.75 0.3255 0.6745
LnTLRate 3.01 1.73 0.3324 0.6676
LnL 3.53 1.88 0.2836 0.7164
LnKWTSupport 1.60 1.27 0.6245 0.3755
LnKPRRoad 1.73 1.32 0.5781 0.4219
LnKNT 4.64 2.15 0.2156 0.7844
----------------------------------------------------
Variable VIF VIF Tolerance Squared
SQRT R-
Collinearity Diagnostics
(obs=504)
> sparency D
. collin LnKNT LnKPRRoad LnKWTSupport LnL LnTLRate LnLTraining_PCI LnEntryCosts LnTran
120
Mô hình không gian 5b xét tác động của các loại hình vốn đầu tư cho giao thông
.
Wald test of spatial terms: chi2(2) = 76.31 Prob > chi2 = 0.0000
/sigma_e .0751891 .0025422 .0703681 .0803405
/sigma_u .3994559 .0367608 .3335302 .4784126
LnKWTSupport .0940846 .0135197 6.96 0.000 .0675864 .1205828
LnKPRRoad .066383 .0288086 2.30 0.021 .0099192 .1228468
M
_cons -1.869242 1.006785 -1.86 0.063 -3.842504 .104019
D .0049861 .0142348 0.35 0.726 -.0229135 .0328857
LnTransparency .0677311 .0413345 1.64 0.101 -.013283 .1487452
LnEntryCosts .0271903 .0387634 0.70 0.483 -.0487845 .1031651
LnLTraining_PCI .2039063 .0456206 4.47 0.000 .1144916 .293321
LnTLRate .1894402 .0333746 5.68 0.000 .1240272 .2548533
LnL .9894797 .075739 13.06 0.000 .8410339 1.137925
LnKWTSupport .008691 .0025166 3.45 0.001 .0037585 .0136236
LnKPRRoad .0222301 .0080301 2.77 0.006 .0064914 .0379687
LnKNT .1584653 .016535 9.58 0.000 .1260572 .1908733
LnGRDP
LnGRDP Coef. Std. Err. z P>|z| [95% Conf. Interval]
Log likelihood = 418.2225 Pseudo R2 = 0.8187
Prob > chi2 = 0.0000
Wald chi2(11) = 3078.17
Obs per group = 8
Group variable: ID Number of groups = 63
Random-effects spatial regression Number of obs = 504
Iteration 1: log likelihood = 418.2225 (backed up)
Iteration 0: log likelihood = 418.2225
Optimizing unconcentrated log likelihood:
Iteration 4: log likelihood = 418.2225
Iteration 3: log likelihood = 418.22249
Iteration 2: log likelihood = 418.21918
Iteration 1: log likelihood = 418.06465
Iteration 0: log likelihood = 411.97728
rescale eq: log likelihood = 411.97728
rescale: log likelihood = 410.17747
improve: log likelihood = 410.17747
initial: log likelihood = 410.17747
Optimizing concentrated log likelihood:
(weighting matrix defines 63 places)
(data contain 63 panels (places) )
(504 observations used)
(504 observations)
> sts LnTransparency D, ivarlag(M: LnKPRRoad LnKWTSupport) re
. spxtregress LnGRDP LnKNT LnKPRRoad LnKWTSupport LnL LnTLRate LnLTraining_PCI LnEntryCo
121
Mô hình không gian 5c xét tác động của các loại hình vốn đầu tư cho giao thông
Wald test of spatial terms: chi2(3) = 87.16 Prob > chi2 = 0.0000
/sigma_e .0740482 .0025042 .0692992 .0791226
/sigma_u .4004152 .0369033 .3342424 .4796887
LnKWaterways -.0378622 .0125085 -3.03 0.002 -.0623784 -.013346
LnKWTSupport .1007749 .0135599 7.43 0.000 .0741981 .1273518
LnKPRRoad .0678596 .0283791 2.39 0.017 .0122376 .1234817
M
_cons -1.954949 1.001699 -1.95 0.051 -3.918243 .0083458
D .0160398 .0145971 1.10 0.272 -.01257 .0446496
LnTransparency .0689114 .0407319 1.69 0.091 -.0109216 .1487445
LnEntryCosts .0341564 .0385257 0.89 0.375 -.0413526 .1096653
LnLTraining_PCI .1988847 .044971 4.42 0.000 .1107432 .2870262
LnTLRate .1847843 .0328987 5.62 0.000 .1203041 .2492646
LnL 1.012552 .0759689 13.33 0.000 .8636561 1.161449
LnKWaterways .0031141 .001853 1.68 0.093 -.0005178 .0067459
LnKWTSupport .0090957 .002487 3.66 0.000 .0042211 .0139702
LnKPRRoad .0215344 .0079198 2.72 0.007 .0060119 .0370568
LnKNT .1566857 .0163152 9.60 0.000 .1247085 .188663
LnGRDP
LnGRDP Coef. Std. Err. z P>|z| [95% Conf. Interval]
Log likelihood = 424.8193 Pseudo R2 = 0.8171
Prob > chi2 = 0.0000
Wald chi2(13) = 3180.96
Obs per group = 8
Group variable: ID Number of groups = 63
Random-effects spatial regression Number of obs = 504
Iteration 1: log likelihood = 424.81933 (backed up)
Iteration 0: log likelihood = 424.81933
Optimizing unconcentrated log likelihood:
Iteration 5: log likelihood = 424.81933
Iteration 4: log likelihood = 424.81933
Iteration 3: log likelihood = 424.81927
Iteration 2: log likelihood = 424.79575
Iteration 1: log likelihood = 420.38519
Iteration 0: log likelihood = 419.25123
rescale eq: log likelihood = 419.25123
rescale: log likelihood = 415.58959
improve: log likelihood = 415.58959
initial: log likelihood = 415.58959
Optimizing concentrated log likelihood:
(weighting matrix defines 63 places)
(data contain 63 panels (places) )
(504 observations used)
(504 observations)
> orce re
> PCI LnEntryCosts LnTransparency D , ivarlag(M: LnKPRRoad LnKWTSupport LnKWaterways) f
. spxtregress LnGRDP LnKNT LnKPRRoad LnKWTSupport LnKWaterways LnL LnTLRate LnLTraining_
122
Mô hình FE 6.1a xét tác động của mật độ đường cao tốc
Mô hình RE 6.2a xét tác động của mật độ đường cao tốc
F test that all u_i=0: F(62, 182) = 213.87 Prob > F = 0.0000
rho .9970208 (fraction of variance due to u_i)
sigma_e .0491398
sigma_u .89895058
_cons 17.01513 2.944732 5.78 0.000 11.20492 22.82533
Y2017 .2292189 .0135673 16.89 0.000 .2024495 .2559882
Y2016 .1576881 .0110858 14.22 0.000 .1358149 .1795613
Y2015 .0808709 .0096071 8.42 0.000 .0619152 .0998266
NaHWDensity .0000997 .0011472 0.09 0.931 -.0021638 .0023632
LnTLRate .0234926 .0435495 0.54 0.590 -.0624341 .1094194
LnL -.0498337 .2216713 -0.22 0.822 -.4872098 .3875424
LnKA .0543935 .0290064 1.88 0.062 -.0028385 .1116255
LnGRDP Coef. Std. Err. t P>|t| [95% Conf. Interval]
corr(u_i, Xb) = 0.2382 Prob > F = 0.0000
F(7,182) = 123.67
overall = 0.1139 max = 4
between = 0.4284 avg = 4.0
within = 0.8263 min = 4
R-sq: Obs per group:
Group variable: ID Number of groups = 63
Fixed-effects (within) regression Number of obs = 252
. xtreg LnGRDP LnKA LnL LnTLRate NaHWDensity Y2015 Y2016 Y2017 if Year >=2014, fe
rho .98220324 (fraction of variance due to u_i)
sigma_e .0491398
sigma_u .36505944
_cons .951871 1.202443 0.79 0.429 -1.404874 3.308616
Y2017 .1763813 .0135787 12.99 0.000 .1497675 .2029952
Y2016 .1228394 .0116333 10.56 0.000 .1000386 .1456401
Y2015 .0621336 .0106247 5.85 0.000 .0413095 .0829576
NaHWDensity .000427 .0012787 0.33 0.738 -.0020791 .0029331
LnTLRate .0871712 .0470137 1.85 0.064 -.004974 .1793164
LnL 1.061359 .0975583 10.88 0.000 .8701486 1.25257
LnKA .1112932 .0314772 3.54 0.000 .0495991 .1729873
LnGRDP Coef. Std. Err. z P>|z| [95% Conf. Interval]
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000
Wald chi2(7) = 868.64
overall = 0.7704 max = 4
between = 0.7707 avg = 4.0
within = 0.7937 min = 4
R-sq: Obs per group:
Group variable: ID Number of groups = 63
Random-effects GLS regression Number of obs = 252
. xtreg LnGRDP LnKA LnL LnTLRate NaHWDensity Y2015 Y2016 Y2017 if Year >=2014, re
123
Kiểm định Lagrange (LM) lựa chọn giữa mô hình POLS và RE
Các kiểm định Hausman lựa chọn giữa mô hình 6.1a và 6.2a
Prob > chibar2 = 0.0000
chibar2(01) = 337.28
Test: Var(u) = 0
u .1332684 .3650594
e .0024147 .0491398
LnGRDP .8518884 .922978
Var sd = sqrt(Var)
Estimated results:
LnGRDP[ID,t] = Xb + u[ID] + e[ID,t]
Breusch and Pagan Lagrangian multiplier test for random effects
. xttest0
see suest for a generalized test
assumptions of the Hausman test;
data fails to meet the asymptotic
= -32.48 chi2 model fitted on these
chi2(7) = (b-B)'[(V_b-V_B)^(-1)](b-B)
Test: Ho: difference in coefficients not systematic
B = inconsistent under Ha, efficient under Ho; obtained from xtreg
b = consistent under Ho and Ha; obtained from xtreg
Y2017 .2292189 .1763813 .0528376 .
Y2016 .1576881 .1228394 .0348488 .
Y2015 .0808709 .0621336 .0187373 .
NaHWDensity .0000997 .000427 -.0003273 .
LnTLRate .0234926 .0871712 -.0636785 .
LnL -.0498337 1.061359 -1.111193 .1990491
LnKA .0543935 .1112932 -.0568997 .
fe re Difference S.E.
(b) (B) (b-B) sqrt(diag(V_b-V_B))
Coefficients
. hausman fe re
124
Kiểm định hiện tượng tự tương quan tại mô hình 6.2a
LM(Var(u)=0,lambda=0) = 337.57 Pr>chi2(2) = 0.0000
Joint Test:
ALM(lambda=0) = 0.29 Pr>chi2(1) = 0.5915
Serial Correlation:
ALM(Var(u)=0) = 12.61 Pr>N(0,1) = 0.0000
Random Effects, One Sided:
ALM(Var(u)=0) = 158.93 Pr>chi2(1) = 0.0000
Random Effects, Two Sided:
Tests:
u .1332684 .36505944
e .0024147 .0491398
LnGRDP .8518884 .922978
Var sd = sqrt(Var)
Estimated results:
v[ID,t] = lambda v[ID,(t-1)] + e[ID,t]
LnGRDP[ID,t] = Xb + u[ID] + v[ID,t]
Tests for the error component model:
. xttest1
125
Mô hình không gian 6b xét tác động của mật độ đường cao tốc trên cả nước
Wald test of spatial terms: chi2(1) = 19.01 Prob > chi2 = 0.0000
/sigma_e .0486567 .002613 .0437956 .0540573
/sigma_u .5016617 .0505008 .4118348 .6110811
NaHWDensity .0272569 .0062511 4.36 0.000 .0150051 .0395088
M
_cons 2.263802 1.528232 1.48 0.139 -.7314771 5.259082
Y2017 .1599086 .0139802 11.44 0.000 .1325079 .1873092
Y2016 .1075812 .0116314 9.25 0.000 .0847841 .1303783
Y2015 .027087 .012912 2.10 0.036 .00178 .052394
NaHWDensity .0012117 .0011437 1.06 0.289 -.0010298 .0034533
LnTLRate .0729456 .0423653 1.72 0.085 -.0100888 .15598
LnL 1.00046 .1152291 8.68 0.000 .7746149 1.226305
LnKA .0753558 .0289358 2.60 0.009 .0186426 .1320691
LnGRDP
LnGRDP Coef. Std. Err. z P>|z| [95% Conf. Interval]
Log likelihood = 213.4850 Pseudo R2 = 0.7251
Prob > chi2 = 0.0000
Wald chi2(8) = 1003.44
Obs per group = 4
Group variable: ID Number of groups = 63
Random-effects spatial regression Number of obs = 252
Iteration 1: log likelihood = 213.48504 (backed up)
Iteration 0: log likelihood = 213.48504
Optimizing unconcentrated log likelihood:
Iteration 4: log likelihood = 213.48504
Iteration 3: log likelihood = 213.48504
Iteration 2: log likelihood = 213.48444
Iteration 1: log likelihood = 213.3326
Iteration 0: log likelihood = 210.91885
rescale eq: log likelihood = 210.91885
rescale: log likelihood = 197.16114
improve: log likelihood = 197.16114
initial: log likelihood = 197.16114
Optimizing concentrated log likelihood:
(weighting matrix defines 63 places)
(data contain 63 panels (places) )
(252 observations used)
(252 observations excluded due to if/in)
(504 observations)
> arlag(M: NaHWDensity) force re
. spxtregress LnGRDP LnKA LnL LnTLRate NaHWDensity Y2015 Y2016 Y2017 if Year>=2014, iv
126
Mô hình RE 7a xét tác động của mật độ đường cao tốc
tại khu vực Quảng Ninh - Huế
rho .96087859 (fraction of variance due to u_i)
sigma_e .05019492
sigma_u .24876357
_cons -.8298606 1.629117 -0.51 0.610 -4.022872 2.363151
Y2017 .1812647 .03042 5.96 0.000 .1216426 .2408868
Y2016 .1310225 .0214179 6.12 0.000 .0890442 .1730008
Y2015 .0703949 .014191 4.96 0.000 .042581 .0982088
NaHWDensity .001407 .0011746 1.20 0.231 -.0008952 .0037092
LnTLRate .3559616 .1177723 3.02 0.003 .1251322 .586791
LnL 1.127176 .1334866 8.44 0.000 .8655475 1.388805
LnKA .1096353 .0515179 2.13 0.033 .008662 .2106085
LnGRDP Coef. Std. Err. z P>|z| [95% Conf. Interval]
Robust
(Std. Err. adjusted for 31 clusters in ID)
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000
Wald chi2(7) = 794.85
overall = 0.8542 max = 4
between = 0.8546 avg = 4.0
within = 0.8347 min = 4
R-sq: Obs per group:
Group variable: ID Number of groups = 31
Random-effects GLS regression Number of obs = 124
> 1, re rob
. xtreg LnGRDP LnKA LnL LnTLRate NaHWDensity Y2015 Y2016 Y2017 if Year >=2014 & ID <=3
127
Mô hình không gian 7b xét tác động của mật độ đường cao tốc
tại khu vực Quảng Ninh - Huế
Wald test of spatial terms: chi2(1) = 15.63 Prob > chi2 = 0.0001
/sigma_e .0464046 .0035202 .0399935 .0538433
/sigma_u .3803183 .0533626 .2888782 .5007023
NaHWDensity .0331916 .0083947 3.95 0.000 .0167382 .049645
M_s001
_cons -.1900516 1.466905 -0.13 0.897 -3.065132 2.685028
Y2017 .1533546 .0232188 6.60 0.000 .1078466 .1988626
Y2016 .0888264 .0203451 4.37 0.000 .0489508 .128702
Y2015 -.0000657 .0235663 -0.00 0.998 -.0462548 .0461234
NaHWDensity .0024312 .0012861 1.89 0.059 -.0000895 .004952
LnTLRate .2268234 .0848163 2.67 0.007 .0605865 .3930603
LnL 1.148207 .1196315 9.60 0.000 .9137333 1.38268
LnKA .0575333 .0329179 1.75 0.081 -.0069845 .1220512
LnGRDP
LnGRDP Coef. Std. Err. z P>|z| [95% Conf. Interval]
Log likelihood = 118.0188 Pseudo R2 = 0.8421
Prob > chi2 = 0.0000
Wald chi2(8) = 757.36
Obs per group = 4
Group variable: ID Number of groups = 31
Random-effects spatial regression Number of obs = 124
Iteration 1: log likelihood = 118.01885 (backed up)
Iteration 0: log likelihood = 118.01885
Optimizing unconcentrated log likelihood:
Iteration 4: log likelihood = 118.01885
Iteration 3: log likelihood = 118.01885
Iteration 2: log likelihood = 118.01511
Iteration 1: log likelihood = 117.50344
Iteration 0: log likelihood = 117.15057
rescale eq: log likelihood = 117.15057
rescale: log likelihood = 109.66494
improve: log likelihood = 109.66494
initial: log likelihood = 109.66494
Optimizing concentrated log likelihood:
(weighting matrix M_s001 created)
(weighting matrix matched 31 places in data)
(you specified -force-)
(weighting matrix defines 63 places)
(data contain 31 panels (places) )
(124 observations used)
(380 observations excluded due to if/in)
(504 observations)
> D <=31, ivarlag(M: NaHWDensity) force re
. spxtregress LnGRDP LnKA LnL LnTLRate NaHWDensity Y2015 Y2016 Y2017 if Year>=2014 & I
128
Mô hình RE 8a xét tác động của mật độ đường cao tốc tại khu vực Đà Nẵng - Cà
Mau
Kiểm định tự tương quan của mô hình 8a
rho .98345161 (fraction of variance due to u_i)
sigma_e .04420078
sigma_u .34074431
_cons 2.179997 1.862795 1.17 0.242 -1.471013 5.831008
Y2017 .1353585 .0191415 7.07 0.000 .0978418 .1728751
Y2016 .0957751 .016648 5.75 0.000 .0631455 .1284046
Y2015 .0454423 .0148843 3.05 0.002 .0162696 .0746149
NaHWDensity -.0007064 .003007 -0.23 0.814 -.0066 .0051873
LnTLRate .0668492 .0551235 1.21 0.225 -.0411908 .1748892
LnL .9188522 .1506282 6.10 0.000 .6236264 1.214078
LnKA .1648978 .0580543 2.84 0.005 .0511135 .2786821
LnGRDP Coef. Std. Err. z P>|z| [95% Conf. Interval]
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000
Wald chi2(7) = 363.80
overall = 0.7577 max = 4
between = 0.7590 avg = 4.0
within = 0.7709 min = 4
R-sq: Obs per group:
Group variable: ID Number of groups = 32
Random-effects GLS regression Number of obs = 128
> 1, re
. xtreg LnGRDP LnKA LnL LnTLRate NaHWDensity Y2015 Y2016 Y2017 if Year >=2014 & ID > 3
LM(Var(u)=0,lambda=0) = 168.43 Pr>chi2(2) = 0.0000
Joint Test:
ALM(lambda=0) = 0.25 Pr>chi2(1) = 0.6175
Serial Correlation:
ALM(Var(u)=0) = 8.82 Pr>N(0,1) = 0.0000
Random Effects, One Sided:
ALM(Var(u)=0) = 77.74 Pr>chi2(1) = 0.0000
Random Effects, Two Sided:
Tests:
u .1161067 .34074431
e .0019537 .04420078
LnGRDP .7482438 .8650109
Var sd = sqrt(Var)
Estimated results:
v[ID,t] = lambda v[ID,(t-1)] + e[ID,t]
LnGRDP[ID,t] = Xb + u[ID] + v[ID,t]
Tests for the error component model:
. xttest1
129
Mô hình không gian 8b xét tác động của mật độ đường cao tốc
tại khu vực Đà Nẵng - Cà Mau
Wald test of spatial terms: chi2(1) = 3.62 Prob > chi2 = 0.0572
/sigma_e .0468297 .0038208 .0399092 .0549502
/sigma_u .5115585 .0868128 .3668126 .7134217
NaHWDensity .0239811 .0126114 1.90 0.057 -.0007369 .048699
M_s001
_cons 5.105834 2.905184 1.76 0.079 -.588221 10.79989
Y2017 .1456491 .0213512 6.82 0.000 .1038015 .1874966
Y2016 .1092266 .0167433 6.52 0.000 .0764104 .1420428
Y2015 .0376157 .0159054 2.36 0.018 .0064417 .0687898
NaHWDensity -.0002737 .0026233 -0.10 0.917 -.0054154 .0048679
LnTLRate .0483648 .0496061 0.97 0.330 -.0488614 .1455911
LnL .7658312 .2026542 3.78 0.000 .3686363 1.163026
LnKA .114014 .0566826 2.01 0.044 .0029181 .2251099
LnGRDP
LnGRDP Coef. Std. Err. z P>|z| [95% Conf. Interval]
Log likelihood = 111.4900 Pseudo R2 = 0.7515
Prob > chi2 = 0.0000
Wald chi2(8) = 407.16
Obs per group = 4
Group variable: ID Number of groups = 32
Random-effects spatial regression Number of obs = 128
Iteration 1: log likelihood = 111.48995 (backed up)
Iteration 0: log likelihood = 111.48995
Optimizing unconcentrated log likelihood:
Iteration 4: log likelihood = 111.48995
Iteration 3: log likelihood = 111.48995
Iteration 2: log likelihood = 111.48972
Iteration 1: log likelihood = 111.39445
Iteration 0: log likelihood = 110.35316
rescale eq: log likelihood = 110.35316
rescale: log likelihood = 101.10642
improve: log likelihood = 101.10642
initial: log likelihood = 101.10642
Optimizing concentrated log likelihood:
(weighting matrix M_s001 created)
(weighting matrix matched 32 places in data)
(you specified -force-)
(weighting matrix defines 63 places)
(data contain 32 panels (places) )
(128 observations used)
(376 observations excluded due to if/in)
(504 observations)
> D > 31, ivarlag(M: NaHWDensity) force re
. spxtregress LnGRDP LnKA LnL LnTLRate NaHWDensity Y2015 Y2016 Y2017 if Year>=2014 & I