At @MotifAnalytics we've adapted LLMs to work on event sequences, and how we are using this approach to generate a far deeper understanding of user experiences and behaviors. Our models don't write SQL for you, they just read your data directly! https://t.co/HR1LjSg7cx
@Carnage4Life Sounds like a nice method to have zero incrementality.
Oh someone in your timeline is already talking about Ant Man? Let's make sure we charge the advertiser for it.
@seanjtaylor@MotifAnalytics If you are looking outside of the business domain in which you are working there are a few intl event datasets
- https://t.co/Ljtibs9VuN
- https://t.co/xdQkgpT90V
@ManagerTactical A club's ownership made some transfers and negotiated salaries. The sum of transfer fees and its wage bill was larger than the number the ownership had in its budget. So now the club refuses to pay the player's wages and transfer fees they agreed to
I ❤️these slides from @mgershoff about Bayesian vs. Frequentist analysis that concludes:
"This is second order issue.... What you are testing, design of experiment, and execution are all an order of magnitude more important than form of analysis."
https://t.co/IoktWx2nrJ
Part 2 of "Casual Forecasting at Lyft" is out! Excited they can share how we built a reliable, holistic model of Lyft's entire business.
Great post by my former teammates @DuaneJRich and @SameerSManek (+ @heng_kuang_ was heavily involved in the project).
https://t.co/De6gCcCNBw
I want to highlight one pattern in the data that surprised us and (we think) warrants further analysis:
We generate pricing policies from different demand models and evaluate profits. Surprisingly(?) profit predictions are almost uncorrelated with statistical measures of fit.
"Optimal Price Targeting" (with @adamnsmith__ & @Iagg11) is now forthcoming @Mrktng_Science:
https://t.co/NkijGNDFXP
We benefitted hugely from the input of @avicgoldfarb, @Guenter_Hitsch, and 2 anonymous referees who helped us improve the paper during the review process.
@Carnage4Life Dysfunctional for the company, but may actually be better for societal innovation. Ideas get diffused quicker through the industry due to employee turnover.