RDocumentation | R package to integrate rdocumentation | Search Engine library

I would like to group my data by state_id and species when I run mice::mice to impute values. I've got it grouped by state_id and results are looking much better than without the bygroup.

mice.impute.bygroup: Groupwise Imputation Function

Edit... improved, working code:

I'm using varImp function from R package caret to get importance of variables. This is my code:

I'm trying to know which observations are active during specific episodes of time

My real objective is to know which Diagnosis (dx) are active during a Pregnancy period (9 months) by a patient. It has to be considered that a patient can have different number of pregnancies along her life and also a different number of diagnosis (the dx can or can't be active ).

I have tried foverlaps like here or here but weren't exactly what I was looking for. The real problem is this and it's well documented but not for R.

Here they make it work for SQL I think. So I hope it's solvable...

I also tried Non-Equi Joins like this but I'm not able to make it work the way I want...

Let's get into it:

I have one DB with patient (id) Diagnosis (dx), the time it was registered (InD_dx) and the time it ended (EndD_dx) like this:

does someone have an idea how to change the number of displayed decimal places when using the ggVennDiagram function (R)?

I'm currently exploring the Lorenz system with Python and R and have noticed subtle differences in the ode packages. odeint from Python and ode both say they use lsoda to calculate their derivatives. However, using the lsoda command for both seems to give far different results. I have tried ode45 for the ode function in R to get something more similar to Python but am wondering why I can't get exactly the same results:

I have a pairwise matrix that I can consider an adjacency matrix of a graph. I am hoping to apply a transitive reduction algorithm to find the graph with the fewest edges but retains the connectivity of the original graph - see image below.

The head of my matrix looks like so:

I am a student interested in hydrology and runoff-simulation. I am using topmodel package in R with huagrahuma dataset, and I want to optimize its parameters by using sceua method in rtop package. I wrote a code, using the example in "rtop package | R Documentation" as a reference. However, I got an error when optimizing the parameters. It says "error in if (sum(mapply(FUN = function(x, y, z) max(y - x, x - z, 0), : missing value where TRUE/FALSE needed". At first, I suspected the error was triggered by NA values which Qobs has. So I converted NA to numerical values but the same error has occurred again. How can I optimize the parameters by SCE-UA method?

Links
topmodel function | R Documentation
https://www.rdocumentation.org/packages/topmodel/versions/0.7.3/topics/topmodel
Tutorial on the use of topmodel in R
https://paramo.cc.ic.ac.uk/topmodel_tutorial
topmodel: Implementation of the Hydrological Model TOPMODEL in R
https://cran.r-project.org/web/packages/topmodel/index.html
rtop package | R Documentation
https://www.rdocumentation.org/packages/rtop/versions/0.5-14
rtop: Interpolation of Data with Variable Spatial Support
https://cran.r-project.org/web/packages/rtop/index.html

In this question / answer from 5 years ago about logLik.lm() and glm(), it was pointed out that code comments in the R stats module suggest that lm() and glm() are both internally calculating some kind of scale or dispersion parameter--presumably one which describes the estimated dispersion of the observation values being predicted by the regression.

This naturally begets another question: if it's truly a real parameter being estimated by the fit algorithm somewhere (or even if it's just some kind of implicit / effective parameter), how do I access this parameter from the resulting fit object?

I've produced a MWE (plus supporting setup / plot code) below:

• Part 1 constructs some simulated input data, which we'll fit to a straight line (implying two fit parameters are expected). Given the question is about a hidden, internally modeled dispersion parameter, I wanted to make sure the fit algorithm is forced to do something interesting, so therefore 10% of the points have been deliberately modeled as outliers. If you understand what's shown in the plot below, then you can probably skip reading this portion of the code.

• Part 2 is the main body of the MWE, illustrating the point that my question is asking about: it runs glm() on the input data and examines some of the results, demonstrating that logLik() claims three parameter estimates, in apparent disagreement with glm() which seems to give two.

• Part 3 just produces a little supplementary figure based on the input data and results. It's only included for completeness & reproducibility; you can probably skip reading it too.

library(survey)

Despite scourging the documentation, I cannot figure out how svytable handles missing values.

As Thomas Lumley mentions in this post svytable does not have a na.rm option. My assumption is that svytable removes the NA's from the calculation of the weighted N's, but I cannot be certain.

Any insight?

Here is an example of my data and method: