Bug Report for ggplot in qgcomp

[chiajungTung]

### Tips for a helpful bug report:

* If you have a question, please don't use this form. Instead, ask on [Stack Overflow](https://stackoverflow.com/) or the [RStudio Community](https://community.rstudio.com/).

* Please include a **minimal reproducible example**, a reprex, to demonstrate the bug. If you've never heard of a reprex, please read ["Make a reprex"](https://www.tidyverse.org/help/#reprex). Do not include session info unless it is explicitly asked for.

* If you can, use one of the built-in datasets or a small toy dataset that exposes the bug. If the bug only occurs on your original data, try to limit the number of rows that are necessary to expose the bug. Share such data by copying `dput()` output rather than an external file.

* Unless the bug is about the theme, labels, scales, or other plot decoration, please omit these from the code.

* Please check whether somebody has reported the same problem in the [issues](https://github.com/tidyverse/ggplot2/issues).

Delete these instructions once you have read them.

---

I found a problem with ...

I expected ...

Here is the code to reproduce the bug:

```r
# copy your code to the clipboard and run:
reprex::reprex()

---

I encountered an issue with the ggplot function in the qgcomp package.

Expected Behavior: I expected the plot(qc_com) function to display the relative weight with a standard bandwidth.

Actual Behavior: However, the bandwidth for the relative weight appears wider than expected.

Here is the code to reproduce the bug:

# Perform quantile g-computation
qc_com <- qgcomp(Com_T_n ~ MP + EP + PP + BP,
                 expnms=c("MP", "EP", "PP", "BP"),
                 data=QWS_dataset, family=gaussian(), q=4)

# Print the results
print(qc_com)

# Plot the results
plot(qc_com)

Please help investigate and resolve this issue. If you require additional information or data to diagnose the problem, please let me know, and I will provide it.

Thank you for your assistance in addressing this bug.

---

[teunbrand]

Can you please boil this down to the issue that arises in the ggplot2 package?

[chiajungTung]

Can you please boil this down to the issue that arises in the ggplot2 package?
Yes

[teunbrand]

I apologise if I've been unclear, but what I meant was: can you show outside the {qgcomp} package what ggplot2 is doing wrong?
Just to reduce the complexity of the problem it is helpful to show a minimal reproducible example.

[teunbrand]

I'm going to close this issue and I'll re-open it if a minimal reprex is posted.