%change price=duration*change yield*country beta This would be giving you the percentage change in price with respect to the foreign currency correct? So we are just assuming exchange rates are constant or being ignored? I guess if the regression (to get country beta) was all done in domestic terms it would give you the percentage change in domestic terms… What do you guys think?
I actually think that exchange rates don’t play any role in here. it’s just the direct relationship of the yield changes in one country and another. That is, exchange rates won’t change the yield in the local country and won’t change the yields in the ldomestic country as well. I guess the perspective is that changes in currencies don’t change yield relationships but maybe change in yields change currencies
yup. remebrr that the beta measures the change of the yld of the fortign bond to change in the yld of domestic bond. If currecny was involved, it’d be even more of a aheadache.
Ok, then this would be the change in foreign price then correct?
yup, with respect to the domestic dont forget that the beat is the sensitivity of the domestic to the froeign.
I would say yes
sbmfj Wrote: ------------------------------------------------------- > yup, with respect to the domestic > dont forget that the beat is the sensitivity of > the domestic to the froeign. I think it is the opposite sbmfj. Domestic rates are the independent variable in the regression with foreign being the dependent. So you take your domestic IR change times the country beta to figure out the foreign rate change then times the duration to figure price change.
yup your right typed to fast change in foreign = b * change in domestic
I think mwvt9 is right you are basically trying to determine what a domestic change in rates will do to your foreign bond price based on sensitivities
I am pretty sure it is the other way around Domestic return % = -1 * (foreign decimal change in yield) * (duration of foreign bond) * (yield beta)
I believe the country beta and the yield are essentially the same concepts. These are just factors that you have to apply to take into account that (for various reasons) a change in yield in a portfolio / in a bond implies a change in yield on the Futures / CTD which is not 1. Can be more, can be less… With regards to the county beta, it is for me a measure of the economies integration. If economies are fully integrated the country beta should be equal to 1. If they are not, I would rather assume the country beta should be below 1 but in the examples, I have seen both cases (below and above 1). Again, my advice (still to be confirmed by some CFAI officials): use the yield / country beta whenever you see it. Otherwise you just assume it is equal to 1. hh, I would agree with you, there is a minus sign in the formula - as with every price / duration formula.
I don’t have the books with me can anybody confirm/infirm the formula?
Precision, I believe that for the country beta calculation, the FX are considered fixed. So any FX movement would come in addition… Example: US rates +25 bps. France beta is 0.8 On top EUR losses 1%. What is the impact on the valuation of a ZC bond valued in EUR with a duration of 5? I would say: In EUR : Delta_price = - 0.25 * 0.8 * 5 In USD: Deta_price = (1 - 0.25 * 0.8 * 5)(1 + 1%) - 1 What do you think? MH
page 124 vol 4 CFAI text from example 16, here it seems that country beta is looking at the impact of a rate change of foreign security on domestic portfolio. Also EOCQ 7 looks at the impact of a domestic interest rate change on a foreign security.
And? So far so good.