Back in May I said that while SciCast hadn’t previously been allowed to pay participants, we were finally running a four week experiment to reward random activities.
They should only reward predictions not comments. Last time people spammed the comment field with inane comments which forced many users to turn off notification and made the comment section less useful.
No, those are not likely to win the accuracy contest. We'll measure the accuracy of a trade as the expected value of its assets given our final prices. Those trades produce almost no net assets gained or lossed, and so have near zero expected value.
I was able to make like $250 or so the last time by always forecasting the current forecast. That seemed to be the fastest way to make a lot of forecasts. Would those forecasts also count as highly accurate? If so, will that confound your results or is there a high volume/low accuracy strategy that I didn't figure out?
How do I find out how I did? There is some ladder but what do those points mean? How many did I have in the beginning? Which forecasts paid off?
This is a great idea but the app is a little messy.
They should only reward predictions not comments. Last time people spammed the comment field with inane comments which forced many users to turn off notification and made the comment section less useful.
No, those are not likely to win the accuracy contest. We'll measure the accuracy of a trade as the expected value of its assets given our final prices. Those trades produce almost no net assets gained or lossed, and so have near zero expected value.
I was able to make like $250 or so the last time by always forecasting the current forecast. That seemed to be the fastest way to make a lot of forecasts. Would those forecasts also count as highly accurate? If so, will that confound your results or is there a high volume/low accuracy strategy that I didn't figure out?