不知道为什么没有直接report pseudo r2, 记得可以通过log pseudolikelihood 算出来，下面这个例子pseudo r2是？并且说明为什么，主要是忘了lnL0是-2402.4524 还是 -3222.8885 了。

. nbreg bidcount a_rival pbeforedumy   acq_adv tgt_adv premium pct_cash Ttdb TlnatD  Tmba tfcfA indumy* yrdumy*, robust cluster(deal_no)

note: indumy1 omitted because of collinearity
note: indumy2 omitted because of collinearity
note: indumy70 omitted because of collinearity
note: indumy73 omitted because of collinearity
note: indumy75 omitted because of collinearity
note: yrdumy1 omitted because of collinearity

Fitting Poisson model:

Iteration 0:   log pseudolikelihood = -2377.8146  
Iteration 1:   log pseudolikelihood = -2377.8137  
Iteration 2:   log pseudolikelihood = -2377.8137  

Fitting constant-only model:

Iteration 0:   log pseudolikelihood = -3222.8885  
Iteration 1:   log pseudolikelihood = -2402.4524  
Iteration 2:   log pseudolikelihood = -2402.4524  

Fitting full model:

Iteration 0:   log pseudolikelihood = -2377.9278  
Iteration 1:   log pseudolikelihood = -2377.8137  
Iteration 2:   log pseudolikelihood = -2377.8137  

Negative binomial regression                      Number of obs   =       2175
Dispersion           = mean                       Wald chi2(79)   =     323.09
Log pseudolikelihood = -2377.8137                 Prob > chi2     =     0.0000

                             (Std. Err. adjusted for 2175 clusters in deal_no)