3 ms·

The only show correlation under an overly simplified model and show odd signs of data massage or p-hacking. For example the TARP paper shows that those with go

The only show correlation under an overly simplified model and show odd signs of data massage or p-hacking. For example the TARP paper shows that those with government connections performed better, but only after 9 months, and we dont know if that was a blip that then reverts back. I've only scanned the paper. Why after 9 months? The usual mechanism of front running the info you would expect the largest difference to be immediate and normalize from there. That is a warning sign for me that these papers aren't very rigorous or their model is too simplistic. The other paper talks of information leakage on large order flow, but that is supposed to happen. When a broker has to transact a large order and he shops it, it would be silly to not expect leakage. The counterparty he goes to now knows about a big order and it will change his perception of things regardless of how many laws you make. With the era of even better algos to sniff out intent, some of this leakage could be from just better data mining tools. Do people expect 100% hidden info to all of the sudden go live and tank a price when a huge trade is printed? Leakage is often wanted to give the market time to adjust.

jnordwick 9y ago
-1 really? This isn't the greatest comment in the world but negative? HN and any financial/economic topics don't mix well at all. Edit: Do people really consider this comment worse than almost every other top-level comment? It currently stands second to last, the last being bitcoin spam. Maybe my view of comment quality is off? (This edit is taking the suggestion from the response below.)
- carapace 9y ago
  Broadly speaking, mentioning downvotes is frowned on here and will get you more downvotes. The exception (as far as I have seen) is asking people to explain their downvotes, e.g. "Downvotes, please explain?" Sometimes that will get you good feedback.
  - jnordwick 9y ago
    Don't care about more downvotes, my average comment score isn't the greatest (I've never even cracked 1900 karma in many many comments over the years). What does interest me is individual comments. It appears to my curiosity about ranking things (like ELO and such in games and sports).
- mancerayder 9y ago
  Some people downvote because they simply didn't like / disagree with an opinion. In my opinion, a bad reason to downvote. Others downvote (like myself, but not in your case) because something doesn't contribute well, harms, or is noise in the conversation. It happens all the time that the former group will do their thing. That's to be expected, really, on an pseudo-anonymous public forum. People sometimes feel like they're voting on opinions and positions within a debate.
- nkurz 9y ago
  This isn't the greatest comment in the world but negative? It's a fine comment. Might be right, might be wrong, but a fine on-topic comment. The only show correlation under an overly simplified model I hope you can appreciate the irony of trying to draw any conclusions from what might have been a single erroneous downvote. At -5 and [flagged], there might be cause for concern. If it happens always, then definitely ask questions. But a single -1 doesn't give much information about group sentiment. Consider that you might be reading too closely into path-dependent noise. In general, almost no one votes, and it's hard to figure out what they are thinking when they do. Possibly someone was bothered by the tiny typo in the first sentence ("The only show"), or perhaps they meant to upvote on a mobile screen and missed. This put the comment at the bottom, and after that it few people read it, and no one else voted. If you are interested, you could ask 'dang' what the actual voting pattern was and he'd probably tell you. So certainly you should wonder after a while if something is wrong with your commenting style if all your comments end up negative, but I wouldn't assume that yet. I'm commenting here in part because I remembered your username from this pleasant exchange you had with BeeOnRope: https://news.ycombinator.com/item?id=16305216 https://news.ycombinator.com/item?id=16305216. Please keep commenting.

jnordwick 9y ago

-1 really? This isn't the greatest comment in the world but negative? HN and any financial/economic topics don't mix well at all. Edit: Do people really consider this comment worse than almost every other top-level comment? It currently stands second to last, the last being bitcoin spam. Maybe my view of comment quality is off? (This edit is taking the suggestion from the response below.)

carapace 9y ago
Broadly speaking, mentioning downvotes is frowned on here and will get you more downvotes. The exception (as far as I have seen) is asking people to explain their downvotes, e.g. "Downvotes, please explain?" Sometimes that will get you good feedback.
- jnordwick 9y ago
  Don't care about more downvotes, my average comment score isn't the greatest (I've never even cracked 1900 karma in many many comments over the years). What does interest me is individual comments. It appears to my curiosity about ranking things (like ELO and such in games and sports).
mancerayder 9y ago
Some people downvote because they simply didn't like / disagree with an opinion. In my opinion, a bad reason to downvote. Others downvote (like myself, but not in your case) because something doesn't contribute well, harms, or is noise in the conversation. It happens all the time that the former group will do their thing. That's to be expected, really, on an pseudo-anonymous public forum. People sometimes feel like they're voting on opinions and positions within a debate.
nkurz 9y ago
This isn't the greatest comment in the world but negative? It's a fine comment. Might be right, might be wrong, but a fine on-topic comment. The only show correlation under an overly simplified model I hope you can appreciate the irony of trying to draw any conclusions from what might have been a single erroneous downvote. At -5 and [flagged], there might be cause for concern. If it happens always, then definitely ask questions. But a single -1 doesn't give much information about group sentiment. Consider that you might be reading too closely into path-dependent noise. In general, almost no one votes, and it's hard to figure out what they are thinking when they do. Possibly someone was bothered by the tiny typo in the first sentence ("The only show"), or perhaps they meant to upvote on a mobile screen and missed. This put the comment at the bottom, and after that it few people read it, and no one else voted. If you are interested, you could ask 'dang' what the actual voting pattern was and he'd probably tell you. So certainly you should wonder after a while if something is wrong with your commenting style if all your comments end up negative, but I wouldn't assume that yet. I'm commenting here in part because I remembered your username from this pleasant exchange you had with BeeOnRope: https://news.ycombinator.com/item?id=16305216 https://news.ycombinator.com/item?id=16305216. Please keep commenting.