PCA on Swap Forwards (Fixed Income Question) - Confused
Hello, everyone... Hope you’re enjoying your weekend. So based on this image above, I know this may be rather intuitive but for me, I'm not quite understanding this concept right now. So the first PC, which is the largest variance (Yield Level) indicates that the yields are increasing as the swap tenor increases but the second PC, second largest variance (Slope) shows the yield decreasing as the tenor increases… Is this not contradicting? Disregard the third PC. Thanks!
This is a linear combination of all tenors. I guess I can't seem to grasp a visual understanding of how these PCs would look as one on the curve. I mean the yield levels are increasing but it's sloping downwards...? If I have to guess, it is indicating a bull flattening curve?
The sign change in PC2 around 3y1y tenor suggests front-end yields moves in the opposite direction from longer-end yields, i.e., bias towards bull flattening or bear steepening. Casual empiricism of the US bond markets over the last 40-50 years would suggest this picture, in part due to Fed easing/hiking cycles which tend to steepen/flatten the curve.
Thanks, Snipez.
Hey Snipez, can you kindly elaborate what you mean by the sign changing in PC2 around 3y1y tenor? From what I see it's a constant upward slope, so are you adding in the level or the curvature factor, as well?
Signs are meaningless in the context of PCA, generally...
Thanks for your input, Martin. Can you elaborate? I'm not sure what signs you are referring to? Sorry if I'm a little slow and I thought change did matter. Like for example, let's say I wanted to take a view on the belly of the curve of a butterfly trade between a 2y10y30y and it looks like there was a cheapening of the 10yr due to a macro factor and I had a specific view and only wanted to be exposed on the curve, so I hedged against the specific drivers of the level and slope. But you're telling me I shouldn't be looking at PCA on an individual basis?
Ah, you know what guys... I'm an idiot... I see the Y-axis shows the change of BPs, so obviously with a mean at 0, nothing changes on the long end but the yields are increasing because it's upward sloping. So from this perspective, we see that bear-flattening seems to be more likely on average than bear-steepening.
But how can the short end of the curve have a negative slope, is it because of the yields are in the negative territory? Possibly due to a driving factor, as Snipez stated, the fed easing?
Corrupti et laboriosam est totam maiores totam cupiditate expedita. Eos vel laborum non rem quia aliquam quo. Aut dicta ut nostrum cupiditate quasi.
Temporibus sint nostrum molestiae sed. Ut et est libero quas incidunt consectetur. Modi aperiam harum dicta qui. Consequatur quo natus qui dolores nam pariatur enim.
Animi sed qui consequatur non ut necessitatibus ipsa voluptas. Laboriosam sint deleniti vero assumenda ut. Consequatur eum ea natus.
Laboriosam nulla ab recusandae voluptate. In et tempore fuga maiores non. Et corporis voluptatem unde voluptatem distinctio autem qui sed.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...