. infile age type dead expose using cda190.raw (4 observations read) . xi: poisson dead i.age i.type, e(expose) i.age Iage_1-2 (naturally coded; Iage_1 omitted) i.type Itype_1-2 (naturally coded; Itype_1 omitted) Poisson regression, normalized by expose Number of obs = 4 Goodness-of-fit chi2(1) = 3.223 Model chi2(2) = 7.618 Prob > chi2 = 0.0726 Prob > chi2 = 0.0222 Log Likelihood = -8.175 Pseudo R2 = 0.3178 ------------------------------------------------------------------------------ dead | Coef. Std. Err. z P>|z| [95% Conf. Interval] ---------+-------------------------------------------------------------------- Iage_2 | 1.220948 .5137939 2.376 0.017 .2139308 2.227966 Itype_2 | -.3298663 .4381649 -0.753 0.452 -1.188654 .5289211 _cons | -6.312098 .5066013 -12.460 0.000 -7.305018 -5.319177 ------------------------------------------------------------------------------ . poisson, ir Poisson regression, normalized by expose Number of obs = 4 Goodness-of-fit chi2(1) = 3.223 Model chi2(2) = 7.618 Prob > chi2 = 0.0726 Prob > chi2 = 0.0222 Log Likelihood = -8.175 Pseudo R2 = 0.3178 ------------------------------------------------------------------------------ dead | IRR Std. Err. z P>|z| [95% Conf. Interval] ---------+-------------------------------------------------------------------- Iage_2 | 3.390401 1.741967 2.376 0.017 1.238537 9.280967 Itype_2 | .7190199 .3150492 -0.753 0.452 .3046311 1.6971 ------------------------------------------------------------------------------ . predict mhat . gen pred = exp(mhat) * expose . sort age type . format mhat pred %6.2f . list age type dead expose mhat pred age type dead expose mhat pred 1. 1 1 4 1259 -6.31 2.28 2. 1 2 1 2082 -6.64 2.72 3. 2 1 7 1417 -5.09 8.72 4. 2 2 9 1647 -5.42 7.28 . gen orate = dead/expose . gen prate = exp(mhat) . format orate prate %7.4f . list age type dead orate prate age type dead orate prate 1. 1 1 4 0.0032 0.0018 2. 1 2 1 0.0005 0.0013 3. 2 1 7 0.0049 0.0062 4. 2 2 9 0.0055 0.0044