Thanks, Rich.
I thought of that, too, but it violates the spirit of my restraints (to
avoid character strings), which I unfortunately did not clearly articulate.
So my apologies for that failure. My concern is that with more complex
model formula, using as.formula, etc. to parse/convert character
I'm still not entirely clear what you want, in particular what you
want to with factor variables ('squaring' them doesn't really seem to
make sense).
Some combination of sprintf/paste/reformulate might do what you want:
## get all second-order terms (doesn't square numeric values)
quad_terms
Arnaud,
I won't comment on other aspects but want to ask how sure you are that your
data is guaranteed to have a single row reflecting a closing price at 18:59:59
exactly?
It may be true for your data source. I note that markets technically close at
4:00 PM, New York time, but many have after-
You might do better posting here:
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
Cheers,
Bert
"An educated person is one who can entertain new ideas, entertain others,
and entertain herself."
On Wed, Apr 2, 2025 at 4:39 PM Arnaud Gaboury
wrote:
> I will try to be more precised. Here is
Hi, Rui and Michael, thank you both for replying.
Yeah, I'm not supposed to know about Chi-squared yet. So far, all of
our work with hypothesis tests has involved creating the sample data,
then resampling it to create a null distribution, and finally computing
p-values.
The prop.test() would work
You could try writing code to write code. However, you need to tell the program
which variables are factor and which are continuous. If you are lucky and all
integer data are factor and all non-integers are continuous, then you can sort
for that. However, in a broad data perspective this is a ba
6 matches
Mail list logo
