Dear Statalist,
I face the following problem:
When calculating a random effects regression model on the influence of Annual Corporate Venture Investments on the firms' innovativeness measured in form of number of patents, the Wald Chi2 and Prob > Chi2 value is missing.
To investigate an inverted U-shaped relationship between these two variables, a squared value of annual CVC investments is utilized.
As soon as I remove the squared value of CVC investments, Wald Chi2 and Prob > Chi2 values are indicated.
Do you have any advice why this problem occurs and how I could solve it.
If you do not have any suggestions, is the missing value a major concern or does the model still maintain its validity?
Kind regards,
Henrik Friede
I face the following problem:
When calculating a random effects regression model on the influence of Annual Corporate Venture Investments on the firms' innovativeness measured in form of number of patents, the Wald Chi2 and Prob > Chi2 value is missing.
To investigate an inverted U-shaped relationship between these two variables, a squared value of annual CVC investments is utilized.
Code:
xtreg number_patents l.annual_cvc l.annual_cvc2 rdi CurrentAssetsTotal log_industry_dynamism
StandardIndustryClassification q_tobin i.fyear, re cluster(gvkey)
Random-effects GLS regression Number of obs = 475
Group variable: gvkey Number of groups = 84
R-squared: Obs per group:
Within = 0.2439 min = 1
Between = 0.0158 avg = 5.7
Overall = 0.1158 max = 14
Wald chi2(19) = .
corr(u_i, X) = 0 (assumed) Prob > chi2 = .
(Std. err. adjusted for 84 clusters in gvkey)
------------------------------------------------------------------------------------------------
| Robust
number_patents | Coefficient std. err. z P>|z| [95% conf. interval]
-------------------------------+----------------------------------------------------------------
annual_cvc_p | 1.01e-06 4.86e-07 2.08 0.037 5.93e-08 1.97e-06
annual_cvc_p2 | -1.01e-15 4.48e-16 -2.26 0.024 -1.89e-15 -1.33e-16
rdi | 209.6591 309.188 0.68 0.498 -396.3382 815.6564
CurrentAssetsTotal | .0006073 .0040613 0.15 0.881 -.0073527 .0085674
log_industry_dynamism | -20.4057 16.01887 -1.27 0.203 -51.8021 10.9907
StandardIndustryClassification | .0665709 .073344 0.91 0.364 -.0771806 .2103224
q_tobin | 27.37014 31.83729 0.86 0.390 -35.0298 89.77008
|
fyear |
2007 | -4.857026 19.56598 -0.25 0.804 -43.20564 33.49158
2008 | 51.24814 117.9224 0.43 0.664 -179.8755 282.3718
2009 | -42.61118 43.22776 -0.99 0.324 -127.336 42.11368
2010 | -19.3521 49.7375 -0.39 0.697 -116.8358 78.13162
2011 | 81.40725 55.00892 1.48 0.139 -26.40824 189.2228
2012 | 140.8808 93.88031 1.50 0.133 -43.12119 324.8829
2013 | 147.7636 95.57647 1.55 0.122 -39.56288 335.09
2014 | 26.19784 65.00353 0.40 0.687 -101.2067 153.6024
2015 | 78.02013 82.21327 0.95 0.343 -83.11492 239.1552
2016 | -10.6446 78.0314 -0.14 0.891 -163.5833 142.2941
2017 | -286.1342 73.27951 -3.90 0.000 -429.7594 -142.5089
2018 | -606.1606 178.5115 -3.40 0.001 -956.0367 -256.2845
2019 | -964.925 199.0749 -4.85 0.000 -1355.105 -574.7454
|
_cons | 73.55724 327.8697 0.22 0.822 -569.0555 716.17
-------------------------------+----------------------------------------------------------------
sigma_u | 601.01498
sigma_e | 409.95135
rho | .68247357 (fraction of variance due to u_i)
As soon as I remove the squared value of CVC investments, Wald Chi2 and Prob > Chi2 values are indicated.
Code:
xtreg number_patents l.annual_cvc rdi CurrentAssetsTotal log_industry_dynamism StandardIndustryClassification q_tobin i.fyear, re cluster(gvkey)
Random-effects GLS regression Number of obs = 475
Group variable: gvkey Number of groups = 84
R-squared: Obs per group:
Within = 0.2359 min = 1
Between = 0.0156 avg = 5.7
Overall = 0.0993 max = 14
Wald chi2(19) = 76.65
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000
(Std. err. adjusted for 84 clusters in gvkey)
------------------------------------------------------------------------------------------------
| Robust
number_patents | Coefficient std. err. z P>|z| [95% conf. interval]
-------------------------------+----------------------------------------------------------------
annual_cvc |
L1. | -1.51e-07 3.19e-07 -0.47 0.636 -7.76e-07 4.75e-07
|
rdi | 239.8759 322.1921 0.74 0.457 -391.609 871.3608
CurrentAssetsTotal | .0013568 .0041526 0.33 0.744 -.0067821 .0094958
log_industry_dynamism | -20.8924 16.49858 -1.27 0.205 -53.22902 11.44421
StandardIndustryClassification | .0659431 .0727926 0.91 0.365 -.0767277 .2086139
q_tobin | 22.6127 31.99808 0.71 0.480 -40.10239 85.32779
|
fyear |
2007 | 2.97971 18.11473 0.16 0.869 -32.5245 38.48392
2008 | 54.23142 117.8298 0.46 0.645 -176.7108 285.1736
2009 | -56.48408 49.60965 -1.14 0.255 -153.7172 40.74905
2010 | -27.87326 52.2187 -0.53 0.593 -130.22 74.47351
2011 | 79.77194 56.71759 1.41 0.160 -31.3925 190.9364
2012 | 140.9612 93.87834 1.50 0.133 -43.03702 324.9593
2013 | 146.3083 94.16647 1.55 0.120 -38.25454 330.8712
2014 | 31.14564 65.17024 0.48 0.633 -96.58569 158.877
2015 | 79.98332 83.16322 0.96 0.336 -83.01361 242.9802
2016 | 18.27598 77.49625 0.24 0.814 -133.6139 170.1658
2017 | -267.867 70.17288 -3.82 0.000 -405.4033 -130.3307
2018 | -586.8568 177.3886 -3.31 0.001 -934.5321 -239.1815
2019 | -931.601 197.4204 -4.72 0.000 -1318.538 -544.6641
|
_cons | 87.81841 326.2587 0.27 0.788 -551.6369 727.2737
-------------------------------+----------------------------------------------------------------
sigma_u | 598.02297
sigma_e | 411.49715
rho | .67866733 (fraction of variance due to u_i)
------------------------------------------------------------------------------------------------
If you do not have any suggestions, is the missing value a major concern or does the model still maintain its validity?
Kind regards,
Henrik Friede
Comment