search

uchicago-computation-workshop / 2018_spring_conference

0 stars 1 forks source link

Wanlin Ji #17

Open bensoltoff opened 6 years ago

rickecon commented 6 years ago

Would help to have a slide the puts your research question clearly in writing. Did you create that table at the end?

jamesallenevans commented 6 years ago

market makers HFT and T-bond futures quoting Morning losses lead to afternoon risk taking? NO; Risk aversion instead This seems intuitive--people could be hedging again further losses and this might indicate absolute thresholds / preferences with respect to losses (too much is too much to lose)

bethbailey commented 6 years ago

Interesting research. I'd love to read your paper when you're finished with it.

How many different market makers are there in this space? Is there any information on the similarity between the algorithms they use?

cxic-mit commented 6 years ago

Very interesting topic! “Morning losses lead to afternoon risk-taking” makes me recall some research in behavioral economics, in which they have different ideas for this issue. For example, the prospect theory tells us that people may be risk averse in the loss domain. I am so interested to see if your results would be different from that.

Neo-Cat commented 6 years ago

@rickecon Sure, it is still a prototype, and the numbers are not in lines, that problem is to be fixed.

Neo-Cat commented 6 years ago

@jamesallenevans It remains part of the null hypothesis that I want to argue against, and it turns out the non-proprietary traders that trades against market makers usually suffers negative positions for around 6 months, but it can get reversed within in-day trading that market makers did. It is best to stay away from gambling!

Neo-Cat commented 6 years ago

@bethbailey Hi, there are about 400 accounted market makers in 10-year treasuries futures market in CBOT but we managed to get 100 of them randomly sampled. Their algorithms did not really plays a role as long as it did not mess up with our assumption, but I doubt whether it is public information since they remains core asset to the market makers.

Neo-Cat commented 6 years ago

@uc-xic Here is a recent behavioral paper based on traders biases with some new findings. https://academic.oup.com/rof/advance-article/doi/10.1093/rof/rfx037/4065206 Hope it could get you some insights on this puzzle!