Open ParkvilleGeek opened 3 years ago
Can you give a full reproducible example? It looks like you might be trying to use the wrong dataset with your fitted model (ie, the model in your get_predicted() call has twin_sex as a predictor instead of cov1 and cov2).
Thanks for the quick reply and apologies for the confusion. I was trying to create a "general" example for the actual data.
Here's a reproducible example
fit_gee <- gee(y ~ x + age + sex, data = df, id = ID) Beginning Cgee S-function, @(#) geeformula.q 4.13 98/01/27 running glm to get initial regression estimate (Intercept) x age sexmale 0.7694870 0.4876350 0.0940889 0.1430764
model_parameters(fit_gee) Parameter | Coefficient | SE | 95% CI | z | p
(Intercept) | 0.77 | 0.30 | [ 0.18, 1.36] | 2.54 | 0.011 x | 0.49 | 0.07 | [ 0.36, 0.62] | 7.24 | < .001 age | 0.09 | 0.02 | [ 0.05, 0.13] | 4.53 | < .001 sex [male] | 0.14 | 0.08 | [-0.02, 0.30] | 1.75 | 0.080
Showing profiled confidence intervals.
vizdata <- visualisation_matrix(df["x"]) vizdata$Predicted <- insight::get_predicted(fit_gee, vizdata) Error in eval(predvars, data, env) : object 'age' not found In addition: Warning message: In predict.lm(object, newdata, se.fit, scale = 1, type = if (type == : calling predict.lm(
) ...
I have attached the df if you have some suggestions.
Many thanks
Hello
Not an issue with the fabulous easystats packages but I'm having problems generating and subsequently plotting estimated values from a GEE model using modelbased functions and see.
The GEE model has two covariates _fitgee <- gee( y ~ x + cov1+ cov2, data = df, id = ID)
parameters can display the model output
Parameter | Coefficient | SE | 95% CI | z | p
(Intercept) | 0.77 | 0.30 | [ 0.18, 1.36] | 2.54 | 0.011 x | 0.49 | 0.07 | [ 0.36, 0.62] | 7.24 | < .001 cov1 | 0.14 | 0.08 | [-0.02, 0.30] | 1.75 | 0.080 cov2 | 0.09 | 0.02 | [ 0.05, 0.13] | 4.53 | < .001
I've been following the vignettes but no success with get_predicted or predict
_> vizdata$Predicted <- get_predicted(fit_gee, vizdata) Error in eval(predvars, data, env) : object 'twinsex' not found In addition: Warning message: In predict.lm(object, newdata, se.fit, scale = 1, type = if (type == : calling predict.lm() ...
Any suggestions?
Many thanks
@ParkvilleGeek