I read the paper yesterday and learned a lot from it, actually. In my view, 4D19's unusual hostility toward this paper says more about his own incentives than anything else, really, as does ODCE's hardwired desire for surprises.
Non-standard errors
-
its a fine paper. and it should be published somewhere. but I am surprised that people are surprised by the takeaways from the paper (i.e., that there is lots of room for p-hacking within commonly accepted designs). my operating assumption is that majority of papers in some fields in finance are p-hacked... either consciously or unconsciously thru ex post justification.
-
No hostility here. In my view, the paper faces two methodological problems: selection bias and lack of a control group. Instead of inferring about my incentives (and gender), why don't you help me understand why these problems are unimportant?
I read the paper yesterday and learned a lot from it, actually. In my view, 4D19's unusual hostility toward this paper says more about his own incentives than anything else, really, as does ODCE's hardwired desire for surprises.
-
Hi 4d19. Can you elaborate on how selection would bias the results of the paper? In my view, self-selection, if anything, reduces dispersion of results as most authors come from the same subfield.
Also, what would be the control group? What is the treatment you have in mind?
No hostility here. In my view, the paper faces two methodological problems: selection bias and lack of a control group. Instead of inferring about my incentives (and gender), why don't you help me understand why these problems are unimportant?
I read the paper yesterday and learned a lot from it, actually. In my view, 4D19's unusual hostility toward this paper says more about his own incentives than anything else, really, as does ODCE's hardwired desire for surprises.
-
There are many Ph.D students (including me) in this project. Do you think publishing this in top journals will be any good?
Not really. You can only say that you were one of the research teams to this project (among the 164 teams). But it can help get visibility to your other papers via Google Scholar, etc.
-
The paper would contribute mostly to finance by emphasizing p-hacking and the replication crisis. But this point has already been made in the sociology literature. The contribution to that field is the focus on peer review and research team quality. But those results turned out rather weak I think.
-
The paper would contribute mostly to finance by emphasizing p-hacking and the replication crisis. But this point has already been made in the sociology literature. The contribution to that field is the focus on peer review and research team quality. But those results turned out rather weak I think.
You should read your own statement back. First, p hacking is a problem as we know from the sociology journal. Second, the results of this paper are not interesting because they are not significant.
-
The paper would contribute mostly to finance by emphasizing p-hacking and the replication crisis. But this point has already been made in the sociology literature. The contribution to that field is the focus on peer review and research team quality. But those results turned out rather weak I think.
You should read your own statement back. First, p hacking is a problem as we know from the sociology journal. Second, the results of this paper are not interesting because they are not significant.
“The results are not interesting because they are not significant”?? What do you even mean?
-
The paper would contribute mostly to finance by emphasizing p-hacking and the replication crisis. But this point has already been made in the sociology literature. The contribution to that field is the focus on peer review and research team quality. But those results turned out rather weak I think.
You should read your own statement back. First, p hacking is a problem as we know from the sociology journal. Second, the results of this paper are not interesting because they are not significant.
“The results are not interesting because they are not significant”?? What do you even mean?
That's what edd0 wrote "The contribution to that field is the focus on peer review and research team quality. But those results turned out rather weak I think." Which is an interesting statement to make after saying that the problem of p-hacking is already well known.
-
That's what edd0 wrote "The contribution to that field is the focus on peer review and research team quality. But those results turned out rather weak I think." Which is an interesting statement to make after saying that the problem of p-hacking is already well known.
Right, so we agree that the paper has little contribution: p-hacking issue is already known, while peer review and research quality results are insignificant? As I see it, only the magnitude result is novel, that non-standard errors are about equally large as standard errors.