The regression model is as follows:
Yijt = b1*insttype j *Rate t-1 +b2*insttype j*Controls t-1+Fixed_effect it +Fixed_effect j+errors ijt
where insttype is a categorical variable with 7 categories and Rate is instrumented by variable GK.
In Stata, I use IVREGHDFE to perform the analysis, given the high dimension of fixed effects in my data. I use the following two methods to identify the effect of different institution type:
Method 1: I select my sample with only the base group and the interested group, then run IVREGHDFE in the selected sample to look at the effect of interest rate on lending for this type of institution w.r.t base group.
Code:
foreach varname in instinvestor financecompany mutualfund insurance ibank othernonbank
foreach varname in insurance{
preserve
keep if `varname' == 1 | nonbank == 0
ivreghdfe logloanamt nonbankGDPl nonbankRGDPfl nonbankCPIl nonbankvixl (nonbankfedfundsl=nonbankGKl), absorb(borrowerid#time lenderid)
}[/CODE]
Method 2: I run IVREGHDFE directly with insttype and try to specify base group as insttype = 0.
Code:
gen inst2 = "" replace inst2 = "bank" if nonbank == 0 replace inst2 = "instinvestor" if nonbank == 1 & instinvestor == 1 replace inst2 = "financecompany" if nonbank == 1 & financecompany == 1 replace inst2 = "mutualfund" if nonbank == 1 & mutualfund == 1 replace inst2 = "insurance" if nonbank == 1 & insurance == 1 replace inst2 = "ibank" if nonbank == 1 & ibank == 1 replace inst2 = "othernonbank" if nonbank == 1 & othernonbank == 1 encode inst2, gen (insttype) ivreghdfe logloanamt i.insttype##(c.lGDP c.lRGDPf c.lcpi c.lvix) (i.insttype##c.lfedfunds=i.insttype##c.lGK), absorb(borrowerid#time lenderid)
note: variable #2 is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
note: variable #9 is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
warning: -ranktest- error in calculating underidentification test statistics;
may be caused by collinearities
note: variable #9 is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
warning: -ranktest- error in calculating underidentification test statistics;
may be caused by collinearities
Here are my sample data and regression results. I've been struggling with the reason for a while. Appreciate your help very much! Thanks in advance!
Array
Array
Code:
* Example generated by -dataex-. To install: ssc install dataex clear input long(borrowerid lenderid) float(time logloanamt) long insttype float(insurance nonbank lfedfunds) double lGK float(lGDP lRGDPf lcpi lvix) 24 1468 163 18.132996 3 0 1 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 5851 163 17.216717 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 5889 163 18.420681 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 5893 163 18.420681 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 6089 163 17.216717 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 6179 163 17.90985 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 6519 163 18.132996 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 6532 163 18.132996 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 6842 213 18.826145 3 0 1 .14 .004999999888241291 1.571948 3.590044 1.5 12.7 24 6895 163 16.118103 3 0 1 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 7828 129 17.141895 1 0 0 3.98 0 2.858653 4.874871 3.3 16.18 24 7842 163 16.81125 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 7845 129 17.141895 1 0 0 3.98 0 2.858653 4.874871 3.3 16.18 24 7861 213 18.826145 1 0 0 .14 .004999999888241291 1.571948 3.590044 1.5 12.7 24 7891 163 16.81125 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 8926 163 17.216717 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 8936 163 17.216717 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 18874 163 17.216717 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 28127 163 17.504398 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 29380 163 16.81125 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 30526 163 17.216717 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 30898 163 16.81125 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 30949 213 18.826145 1 0 0 .14 .004999999888241291 1.571948 3.590044 1.5 12.7 24 32764 163 16.81125 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 36078 163 17.504398 1 0 0 6.52 .004999999888241291 4.07533 .5352923 1.8 20.57 24 38939 213 18.826145 1 0 0 .14 .004999999888241291 1.571948 3.590044 1.5 12.7 24 77486 213 18.826145 7 0 1 .14 .004999999888241291 1.571948 3.590044 1.5 12.7 24 139639 213 18.826145 1 0 0 .14 .004999999888241291 1.571948 3.590044 1.5 12.7 26 5851 165 17.034386 1 0 0 5.31 -.19499000906944275 2.30971 -1.1358403 1.8 28.64 26 5889 149 18.420681 1 0 0 5.39 .03999999910593033 4.314064 2.6073704 1.9 22.14 26 5889 165 17.50439 1 0 0 5.31 -.19499000906944275 2.30971 -1.1358403 1.8 28.64 26 5893 165 17.034386 1 0 0 5.31 -.19499000906944275 2.30971 -1.1358403 1.8 28.64 26 5893 205 20.18095 1 0 0 .14 -.004999999888241291 1.930627 -.9583018 1.1 17.74 26 5893 222 20.435184 1 0 0 .13 -.019999999552965164 3.350027 2.9985785 1.3 18.23 26 5893 232 17.833694 1 0 0 1.3 0 2.795257 3.545638 1.7 11.04 26 5952 149 18.420681 1 0 0 5.39 .03999999910593033 4.314064 2.6073704 1.9 22.14 26 6050 165 17.50439 1 0 0 5.31 -.19499000906944275 2.30971 -1.1358403 1.8 28.64 26 6050 206 18.420681 1 0 0 .09 -.009999999776482582 1.721503 2.890711 1.5 16.52 26 6123 227 18.988264 1 0 0 .4 -.009999999776482582 1.556712 2.1868734 1.7 13.29 26 6123 238 19.27083 1 0 0 2.38 0 2.27832 2.0138185 1.6 15.08 26 6127 206 17.727533 1 0 0 .09 -.009999999776482582 1.721503 2.890711 1.5 16.52 26 6127 222 18.064005 1 0 0 .13 -.019999999552965164 3.350027 2.9985785 1.3 18.23 26 6127 238 18.064005 1 0 0 2.38 0 2.27832 2.0138185 1.6 15.08 26 6179 165 17.909855 1 0 0 5.31 -.19499000906944275 2.30971 -1.1358403 1.8 28.64 26 6443 165 17.909855 1 0 0 5.31 -.19499000906944275 2.30971 -1.1358403 1.8 28.64 26 6532 165 18.133 1 0 0 5.31 -.19499000906944275 2.30971 -1.1358403 1.8 28.64 26 6541 205 18.538464 1 0 0 .14 -.004999999888241291 1.930627 -.9583018 1.1 17.74 26 6541 206 18.826145 1 0 0 .09 -.009999999776482582 1.721503 2.890711 1.5 16.52 26 6541 222 20.366304 1 0 0 .13 -.03999999910593033 3.350027 2.9985785 1.3 18.23 26 6541 232 17.833694 1 0 0 1.3 0 2.795257 3.545638 1.7 11.04 26 6541 238 19.11383 1 0 0 2.38 0 2.27832 2.0138185 1.6 15.08 26 6542 206 17.727533 1 0 0 .09 -.009999999776482582 1.721503 2.890711 1.5 16.52 26 6542 222 20.02972 1 0 0 .13 -.019999999552965164 3.350027 2.9985785 1.3 18.23 26 6542 232 17.833694 1 0 0 1.3 0 2.795257 3.545638 1.7 11.04 26 6542 238 19.708534 1 0 0 2.38 0 2.27832 2.0138185 1.6 15.08 26 6842 206 18.420681 3 0 1 .09 -.009999999776482582 1.721503 2.890711 1.5 16.52 26 6842 232 17.833694 3 0 1 1.3 0 2.795257 3.545638 1.7 11.04 26 6842 238 19.708534 3 0 1 2.38 0 2.27832 2.0138185 1.6 15.08 26 6884 131 16.118206 5 1 1 3.22 -.2800000011920929 3.66535 4.0120754 2.9 14.28 26 6885 131 17.172213 5 1 1 3.22 -.2800000011920929 3.66535 4.0120754 2.9 14.28 26 6885 133 16.108046 5 1 1 3.07 -.019999999552965164 3.320909 .6712845 2.7 12.53 26 6994 133 16.300417 5 1 1 3.07 -.019999999552965164 3.320909 .6712845 2.7 12.53 26 7033 131 15.201517 5 1 1 3.22 -.2800000011920929 3.66535 4.0120754 2.9 14.28 26 7034 131 15.201517 5 1 1 3.22 -.2800000011920929 3.66535 4.0120754 2.9 14.28 26 7052 131 16.118431 5 1 1 3.22 -.2800000011920929 3.66535 4.0120754 2.9 14.28 26 7053 131 16.523651 5 1 1 3.22 -.2800000011920929 3.66535 4.0120754 2.9 14.28 26 7061 133 15.895077 5 1 1 3.07 -.019999999552965164 3.320909 .6712845 2.7 12.53 26 7062 131 15.519858 5 1 1 3.22 -.2800000011920929 3.66535 4.0120754 2.9 14.28 26 7063 131 13.81511 5 1 1 3.22 -.2800000011920929 3.66535 4.0120754 2.9 14.28 26 7094 131 15.201517 5 1 1 3.22 -.2800000011920929 3.66535 4.0120754 2.9 14.28 26 7161 131 13.81456 5 1 1 3.22 -.2800000011920929 3.66535 4.0120754 2.9 14.28 26 7169 131 14.509032 5 1 1 3.22 -.2800000011920929 3.66535 4.0120754 2.9 14.28 26 7195 131 13.81511 5 1 1 3.22 -.2800000011920929 3.66535 4.0120754 2.9 14.28 26 7827 238 19.708534 1 0 0 2.38 0 2.27832 2.0138185 1.6 15.08 26 7828 206 17.727533 1 0 0 .09 -.009999999776482582 1.721503 2.890711 1.5 16.52 26 7828 227 18.988264 1 0 0 .4 -.009999999776482582 1.556712 2.1868734 1.7 13.29 26 7828 232 17.833694 1 0 0 1.3 0 2.795257 3.545638 1.7 11.04 26 7828 238 19.708534 1 0 0 2.38 0 2.27832 2.0138185 1.6 15.08 26 7832 222 18.064005 1 0 0 .13 -.019999999552965164 3.350027 2.9985785 1.3 18.23 26 7845 222 18.826145 1 0 0 .13 -.019999999552965164 3.350027 2.9985785 1.3 18.23 26 7845 227 18.988264 1 0 0 .4 -.009999999776482582 1.556712 2.1868734 1.7 13.29 26 7845 232 17.833694 1 0 0 1.3 0 2.795257 3.545638 1.7 11.04 26 7845 238 19.708534 1 0 0 2.38 0 2.27832 2.0138185 1.6 15.08 26 7859 227 18.988264 1 0 0 .4 -.009999999776482582 1.556712 2.1868734 1.7 13.29 26 7859 232 17.833694 1 0 0 1.3 0 2.795257 3.545638 1.7 11.04 26 7859 238 19.708534 1 0 0 2.38 0 2.27832 2.0138185 1.6 15.08 26 7861 205 18.538464 1 0 0 .14 -.004999999888241291 1.930627 -.9583018 1.1 17.74 26 7861 206 18.420681 1 0 0 .09 -.009999999776482582 1.721503 2.890711 1.5 16.52 26 7861 222 20.435184 1 0 0 .13 -.019999999552965164 3.350027 2.9985785 1.3 18.23 26 7861 227 18.988264 1 0 0 .4 -.009999999776482582 1.556712 2.1868734 1.7 13.29 26 7861 232 17.833694 1 0 0 1.3 0 2.795257 3.545638 1.7 11.04 26 7861 238 19.708534 1 0 0 2.38 0 2.27832 2.0138185 1.6 15.08 26 7878 149 18.420681 1 0 0 5.39 .03999999910593033 4.314064 2.6073704 1.9 22.14 26 7878 165 18.064005 1 0 0 5.31 -.19499000906944275 2.30971 -1.1358403 1.8 28.64 26 7878 206 18.133 1 0 0 .09 -.009999999776482582 1.721503 2.890711 1.5 16.52 26 7878 222 18.515991 1 0 0 .13 -.019999999552965164 3.350027 2.9985785 1.3 18.23 26 7878 238 19.708534 1 0 0 2.38 0 2.27832 2.0138185 1.6 15.08 26 7886 206 17.727533 1 0 0 .09 -.009999999776482582 1.721503 2.890711 1.5 16.52 26 7909 206 17.727533 1 0 0 .09 -.009999999776482582 1.721503 2.890711 1.5 16.52 26 7909 232 17.833694 1 0 0 1.3 0 2.795257 3.545638 1.7 11.04 end format %tq time label values insttype insttype label def insttype 1 "bank", modify label def insttype 3 "ibank", modify label def insttype 5 "insurance", modify label def insttype 7 "othernonbank", modify label var borrowerid "BorrowerCompanyID" label var lenderid "CompanyID" label var insurance "(mean) insurance" label var nonbank "(mean) nonbank" label var lfedfunds "(lastnm) lfedfunds" label var lGK "(sum) lGK" label var lGDP "(lastnm) lGDP" label var lRGDPf "(lastnm) lRGDPf" label var lcpi "(lastnm) lcpi" label var lvix "(lastnm) lvix"
0 Response to IVREGHDFE: Collinearity with fixed effects
Post a Comment