July 28th, 2022, 8:41 pm
I'm with Cuch on this one. There are probably hundreds of resarch papers online that "demonstrate the superiority of [ML model] to forecast the yield curve". FWIW I was discussing recently with a fixed-income data scientist (I'm not sure of his official job title) working with a well-known Middle East sovereign fund (you know who I mean). The guy post short pieces of research on his blog from time to time (don't ask me I won't disclose much. you have to find find yourself). I asked him Hey do you do those things in real life? His answer was a fat NO.