You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
Monetizing the prediction is the job of portfolio construction (where you trade off alpha vs. risk, costs and constraints).
The backtest is just to make sure that all these parts are working together as expected. In many ways a slow backtest can be a feature, because it discourages you from running hundreds of simulations and optimizing based on the results!
Idk how much you have seen this, but it also seems weird to me on how people don't consider what backtest is supposed to represent. For example, running a backtest of L/S momentum variation.... where each side is 100 names but then in execution they L/S 5 names on each side.
It seems like there needs to be work done on analyzing the results of a backtest itself vs portfolio composition. Obviously there is the classic factor attribution and such, but we should think more about how the composition of same strategy effects the pefofmrnace.
I would simply use BTST<GO> for all my back testing needs.
How would you study event based seasonality, without predominantly looking at backtest results? Say, you think that the market rallies before FOMC. It's straightforward to pick a 1d lookahead and run some statistical test, but that doesn't tell you when you should put it on.
You don't have a fundamental reason that the rally should start 1 hour vs 1 day before, but you do think there's an reason for it to happen eventually. Optimizing a backtest metric seems reasonable, because of the simplicity of the model, but I'd be interested in other opinions
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
