Hi there, I am working with a very large panel dataset and would like to use the ConleySEs function that you've written. However, I'm not sure why I keep getting this error message:

Error in XeeXhC_Lg(cbind(lat, lon), cutoff, X, e, n1, k, kernel, dist_fn) :
matrix multiplication: incompatible matrix dimensions: 1x7 and 1x1085973

Here is the traceback:
6. | stop(structure(list(message = "matrix multiplication: incompatible matrix dimensions: 1x7 and 1x1085973", call = XeeXhC_Lg(cbind(lat, lon), cutoff, X, e, n1, k, kernel, dist_fn), cppstack = NULL), class = c("std::logic_error", "C++Error", "error", "condition")))
5. | XeeXhC_Lg(cbind(lat, lon), cutoff, X, e, n1, k, kernel, dist_fn) at iterate-obs-function.R#29
4. | iterateObs(sub_index = t, type = "spatial", cutoff = dist_cutoff)
3. FUN(X[[i]], ...)
2. | lapply(timeUnique, function(t) iterateObs(sub_index = t, type = "spatial", cutoff = dist_cutoff))

1. | ConleySEs(reg = m, unit = "stdyrID", time = "year", lat = "lat", lon = "lon", dist_cutoff = 0.05, lag_cutoff = 2, cores = 1, verbose = FALSE)
   -- | --

I'm having trouble using the R code to produce standard errors in a cross section and more generally in cases where one does not include a unit fixed effect.

There's a workaround for the hard-coded requirement of two fixed by simply including a constant as one fixed effect. However, the

" # Renaming variables:
orig_names <- c(unit, time, lat, lon)
new_names <- c("unit", "time", "lat", "lon")
setnames(dt, orig_names, new_names)"

lines produce an error as the "unit" variable is not in the model. Can you recommend a workaround? thanks!