andrewraim
commented
4 years ago Hello,
It sounds like you may be calling the nu function on an object which was not output from the glm.cmp function. Can you provide a minimal-but-complete example that I can use to replicate the error? Also, are you using the latest version of the package from CRAN, or something else?
Here is an example where the nu function works as intended. I used COMPoissonReg version 0.7.0.
R> library(COMPoissonReg)
R> data(freight)
R> cmp.out = glm.cmp(broken ~ transfers, data=freight)
R> print(cmp.out)
CMP coefficients
Estimate SE z.value p.value
X:(Intercept) 13.8330 6.2434 2.2156 0.02672
X:transfers 1.4847 0.6895 2.1534 0.03129
S:(Intercept) 1.7553 0.4493 3.9066 9.36e-05
--
Transformed intercept-only parameters
Estimate SE
nu 5.7853 2.5994
--
Chi-squared test for equidispersion
X^2 = 9.1048, df = 1, p-value = 2.5494e-03
--
Elapsed Sec: 0.03 Sample size: 10 SEs via Hessian
LogLik: -18.6449 AIC: 43.2898 BIC: 44.1975
Optimization Method: L-BFGS-B Converged status: 0
Message: CONVERGENCE: REL_REDUCTION_OF_F <= FACTR*EPSMCH
R> coef(cmp.out)
X:(Intercept) X:transfers S:(Intercept)
13.833003 1.484698 1.755315
R> nu(cmp.out)
[1] 5.78527 5.78527 5.78527 5.78527 5.78527 5.78527 5.78527 5.78527 5.78527 5.78527
Hello, I was trying to fit under dispersed count data using "nu" function but got an error " no applicable method for 'nu' applied to an object of class "c('glm', 'lm')"". please suggest me how to solve the issue. Thank you