Hmisc
Hmisc copied to clipboard
harrelfe
Harrell Miscellaneous
I believe this caused the package `neuropsychology` to fail against dplyr 1.0.0 because a data frame with additional class `describe` is `[` and this method only keeps "describe" as the...
Original code has two problems: 1. When normwt = FALSE and weight sum < 1.  This line goes wrong. The indexing becomes descending. 2. It sees the lowest index...
Prof Harrell, I am getting the wrong output when using summaryM for 2 grouping variables. Please find attached the code ```{r, warning=F, message=F} library(Hmisc) getHdata(pbc) s2
wtd.quantile
I noticed a very surprising result of the wtd.quantile function. If the weights are not integers, there seems to be an issue: `> require(Hmisc)` `> data weights wtd.quantile(data, weights=weights, probs=0.5)`...
Hi In a specific case `Hmisc::largest.empty(x, y, method="area")` does not return the largest empty rectangle (code example [Hmisc_largest.empty_example.txt](https://github.com/harrelfe/Hmisc/files/5364263/Hmisc_largest.empty_example.txt)):  The area of `Hmisc::largest.empty` (blue) is 1078482. The area of...
Within `R/latex.s`, `format.df(x, ...)` relies on detecting the type of its argument `x` to determine how to access its attributes. Here are the key lines from that method: https://github.com/harrelfe/Hmisc/blob/6d9bd1f5bc254b0f95cc6fdfbb0e1ca4f761a06f/R/latex.s#L140-L144 The...
Example: wtd.var(c(NA,NA,NA), weights=1:3, method='ML') Error in stats::cov.wt(cbind(x), weights, method = "ML") : weights must be non-negative and not all zero raises an error, while wtd.var(c(NA,NA,NA), method='ML') # [1] NA does...
After installing the package `Hmisc` and loading it I get following error: ``` > install.packages("Hmisc") > library("Hmisc") Error: package or namespace load failed for ‘Hmisc’ in loadNamespace(j
I usually create the typical ''Table 1'' (basic patient demographics, potentially with some stratification like gender) with the `summary` methods of `Hmisc` (eg. `summaryM`, `summary.formula`). Their `print` method works nicely...