I would like to hear your opinion on the results I get from the reverse causality tests I performed (via the commands pvar & pvargranger). I am using Stata16. My panel dataset contains 17 countries (with t=6, 1995 - 2020), and a total of 102 observations. Here is the data of the three variables I want to perform a reverse causality test (lngdp , as my DV, and z_exp_diss and z_rd as my IVs) :
Code:
* Example generated by -dataex-. For more info, type help dataex clear input float(ln_gdppc z_exp_diss z_rd z_labourcost) 7.722362 1.3693258 -1.0572331 -3.1591444 7.390948 1.3693258 -.8968285 -1.1011491 8.268708 1.0252694 -1.0343182 -.70354 8.826501 .5967307 -.7364238 .4561532 8.861625 .3057935 .13434455 1.313958 9.193006 -.03913696 -.11771997 2.2822099 8.4923935 .6811494 .13434455 -1.3257246 8.484998 .6758959 .13434455 -.423741 9.261784 .24520718 -.071890056 -.07031069 9.543186 .08851427 -.3468695 .4561532 9.374306 -.15563676 -.11771997 .18739893 9.611536 -.3543932 .50098383 .24630398 9.633082 1.6185626 -1.584277 -.788216 9.574174 1.242349 -1.5155323 -.5673221 10.125 .7579433 -1.194723 .02172846 10.342505 .644174 -1.0343182 .4561532 10.057654 1.7487774 -.9426584 .18739893 10.23489 1.7039355 -.5989341 .2757565 8.669764 -.9285728 -.026060145 -1.2152777 8.704343 -1.470508 .50098383 -.3942885 9.505296 -1.584609 .6155586 .11744916 9.901489 -1.3211894 1.0051129 .4561532 9.788621 -1.395671 2.3570952 .6255053 10.064526 -1.2370006 2.402925 1.3176396 8.0490465 .7375261 -.7364238 -1.796965 8.312864 1.1607805 -.6676789 -1.152691 9.250176 .45456 .06559968 -.6777691 9.601762 .2032704 1.555072 .4561532 9.771226 .2759098 1.3030072 1.24769 10.074213 -.004973041 1.6467316 2.0981317 9.469571 1.4492502 -1.0801481 -1.2410486 9.396235 1.1072044 -.7593387 -.6888137 10.023567 .699696 -.7135088 -.0040425034 10.200541 .9931969 -.6676789 .4561532 9.807405 1.1554782 .18017446 .13217543 9.882394 1.0336165 .8676231 .1910805 8.410656 -.06580974 -.4156144 -2.0362668 8.439076 -.0849603 -.23229474 -.9097077 9.323721 -.3819059 .06559968 .014365326 9.487337 -.4923444 .5468137 .4561532 9.4499 -1.281711 1.0280278 .7838126 9.725068 -1.3347082 1.348837 1.4133604 9.936175 -1.348037 .08851463 -.6740875 9.907858 -1.1452198 .24891932 -.4568751 10.37484 -1.1431822 .34057915 .02172846 10.491288 -1.1320345 .7530484 .4561532 10.316598 -1.0410514 1.0280278 .6070974 10.411156 -.1259437 1.2800924 .7433153 7.752862 .9413027 -1.0572331 -1.679155 8.118978 1.6185786 -1.0343182 -1.1858251 8.932632 1.0744585 -.8280836 -.6740875 9.339922 .4698162 -.6447639 .4561532 9.530582 .6044782 -.621849 1.1851033 9.788576 .54994786 -.57601905 2.1680813 7.681465 .7674782 -1.0572331 -1.263138 8.099624 1.209263 -.6905938 -.3795622 8.968876 .8577559 -.3239546 -.09240008 9.391884 .7559402 -.2552097 .4561532 9.565089 .51060766 .34057915 .7838126 9.883381 .1518277 .24891932 2.002411 9.195986 1.7571434 -1.4697024 -.3169756 9.252665 2.0901248 -1.4697024 -.3169756 9.67333 1.0573235 -.8280836 .1027229 9.989627 1.4017817 -.6905938 .4561532 10.12349 1.2024 -.3926994 .780131 10.302955 .8918662 -.6905938 1.4354497 8.21252 .29601064 -.621849 -1.288909 8.412156 -.5381761 -.57601905 .018046891 8.9898815 -1.0286179 -.7593387 .007002194 9.442484 -1.1793464 -.3926994 .4561532 9.439744 -1.034704 .24891932 .5997343 9.660938 -1.0364012 .9821979 1.008388 9.374274 .2080747 -.8509985 -.7256294 9.349907 -.3067259 -.3926994 -.2102102 9.840181 -.5954473 -.3010396 .24630398 10.021213 -.16994613 1.486327 .4561532 9.86487 -.12019895 .7988783 .27207494 10.05415 -.4888581 1.1655176 .55923706 7.408698 1.1293633 -.3010396 -3.1333735 7.414517 1.1625406 -1.194723 -2.0878088 8.437701 .7149008 -1.1030631 -.7072216 9.013605 -.4342212 -.9884883 .4561532 9.101546 -.68262 -.9197434 .3162537 9.4664955 -1.0684882 -.9426584 1.6416174 8.480348 -.1563294 .019769765 -1.2263223 8.596586 -.7502491 -.57601905 -.39797005 9.366126 -1.1660045 -.9197434 .09904134 9.726201 -.6055823 -.6447639 .4561532 9.699594 -.6734886 .6155586 .6181421 9.866112 -.54155046 -.14063492 1.2845055 9.280841 -.9585403 1.371752 -1.425127 9.23027 -.9140266 1.0738577 -.8066239 9.803607 -1.2356193 1.2113475 -.1549867 10.065162 -1.439919 2.65499 .4561532 9.946631 -1.4153608 2.998714 .4451085 10.16378 -1.3464882 2.6320746 .8206282 9.646785 -1.071029 -.27812466 -.7366741 9.596491 -1.164435 -.026060145 -.4090147 10.18185 -1.296438 .4780689 .003320628 10.32557 -1.1727965 1.0738577 .4561532 end
Now, before doing the reverse causality test, I run a fixed effects model and a pooled OLS model. (The command for the fe model was:
Code:
xtreg ln_gdppc z_exp_diss z_rd z_labourcost z_invest z_hc, fe cluster(country)
Code:
pvar ln_gdppc z_exp_diss z_rd pvargranger
Code:
Panel vector autoregresssion GMM Estimation Final GMM Criterion Q(b) = 8.91e-33 Initial weight matrix: Identity GMM weight matrix: Robust No. of obs = 68 No. of panels = 17 Ave. no. of T = 4.000 ------------------------------------------------------------------------------ | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- ln_gdppc | ln_gdppc | L1. | .0597939 .8042858 0.07 0.941 -1.516577 1.636165 | z_exp_diss | L1. | -.9220663 .8075951 -1.14 0.254 -2.504924 .660791 | z_rd | L1. | .2887812 .3659468 0.79 0.430 -.4284613 1.006024 -------------+---------------------------------------------------------------- z_exp_diss | ln_gdppc | L1. | .0740402 .769195 0.10 0.923 -1.433554 1.581635 | z_exp_diss | L1. | 1.193323 .7442247 1.60 0.109 -.2653302 2.651977 | z_rd | L1. | .161285 .3361554 0.48 0.631 -.4975675 .8201375 -------------+---------------------------------------------------------------- z_rd | ln_gdppc | L1. | -.0983861 1.350602 -0.07 0.942 -2.745517 2.548745 | z_exp_diss | L1. | -.3242725 1.424901 -0.23 0.820 -3.117026 2.468481 | z_rd | L1. | .9076021 .6355367 1.43 0.153 -.3380269 2.153231 ------------------------------------------------------------------------------ Instruments : l(1/1).(ln_gdppc z_exp_diss z_rd)
Code:
pvargranger panel VAR-Granger causality Wald test Ho: Excluded variable does not Granger-cause Equation variable Ha: Excluded variable Granger-causes Equation variable +------------------------------------------------------+ | Equation \ Excluded | chi2 df Prob > chi2 | |----------------------+-------------------------------| |ln_gdppc | | | z_exp_diss | 1.304 1 0.254 | | z_rd | 0.623 1 0.430 | | ALL | 1.688 2 0.430 | |----------------------+-------------------------------| |z_exp_diss | | | ln_gdppc | 0.009 1 0.923 | | z_rd | 0.230 1 0.631 | | ALL | 2.879 2 0.237 | |----------------------+-------------------------------| |z_rd | | | ln_gdppc | 0.005 1 0.942 | | z_exp_diss | 0.052 1 0.820 | | ALL | 0.770 2 0.681 | +------------------------------------------------------+
Maybe somebody can help me with the interpretation. Thank you so much !
Warm regards
Valentina
0 Response to Problem with pvar and pvargranger command & interpretation
Post a Comment