# non parametric ANCOVA with 2 factors and 2 covariates (npsm package)

Hi, and welcome!

There are no `stupid questions` here. We're a community of learners from beginners through intermediate and advanced, with occasional visits from on high--some of the top luminaries in the `R` universe.

One of the things that allows us to be helpful is reproducible examples, called a reprex . In the case of the code snippet, `dataset` is not in the namespace by default. Representative data in the same form (even from a built-in data set) makes for more and better answers.

In this case, though, running the example from `help(package = Rfit, rfit)` illustrates the problem:

``````library(Rfit)
data(baseball)
data(wscores)
fit<-rfit(weight~height,data=baseball)
summary(fit)
#> Call:
#> rfit.default(formula = weight ~ height, data = baseball)
#>
#> Coefficients:
#>               Estimate Std. Error t.value   p.value
#> (Intercept) -228.57143   56.14659 -4.0710  0.000146 ***
#> height         5.71429    0.76018  7.5171 4.373e-10 ***
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> Multiple R-squared (Robust): 0.4418605
#> Reduction in Dispersion Test: 45.125 p-value: 0
``````

Created on 2020-01-03 by the reprex package (v0.3.0)

Then, take a look at the `rfit` function `signature`

Default S3 method:
rfit(formula, data, subset, yhat0 = NULL,
scores = Rfit::wscores, symmetric = FALSE, TAU = "F0", ...)

`subset` is optional, but it needs both `formula` and `data`. The error thrown by the code snippet reflects the missing `data` argument, which would tell `rfit` where to find `xfull.`