请问您有什么建议吗？我对R进行了felm回归，并对利息变量的系数进行了回归，使用IV（税率变量的滞后值）后，Tax变为NA。

felm(formula = log(IFDI) ~ Tax + log(`Monthly Earnings`) + log(GDP) | Economy + Year | (Tax | GDP ~ lag(Tax,3) + lag(GDP,1)) | Economy,data = joined3) %>%
  summary(Model9,Robust = TRUE)

谢谢。

Residuals:
     Min       1Q   Median       3Q      Max 
-0.72916 -0.07161  0.00000  0.07114  0.45628 

Coefficients:
                          Estimate Cluster s.e. t value Pr(>|t|)    
Tax                             NA    0.000e+00      NA       NA    
log(`Monthly Earnings`) -3.206e-01    3.012e-01  -1.064 0.290229    
log(GDP)                 1.323e+00    3.525e-01   3.752 0.000326 ***
`Tax(fit)`               2.133e-02    9.229e-03   2.311 0.023345 *  
`GDP(fit)`              -5.952e-08    7.802e-08  -0.763 0.447693    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.1676 on 82 degrees of freedom
  (358 observations deleted due to missingness)
Multiple R-squared(full model): 0.9921   Adjusted R-squared: 0.9884 
Multiple R-squared(proj model): 0.4217   Adjusted R-squared: 0.1466 
F-statistic(full model,*iid*):  265 on 39 and 82 DF,p-value: < 2.2e-16 
F-statistic(proj model): 7.129 on 5 and 20 DF,p-value: 0.0005641 
F-statistic(endog. vars):6.484 on 2 and 20 DF,p-value: 0.00675