YIELD Function return a strange number

[massanpc]

Assumptions:

• settlement: Jul 31st, 2020
• maturity: Sep 30th, 2020
• rate: 2%
• pr: 100.163338941426
• yld: 1%
• redemption: 100
• frequency: 4
• basis: 2

Based on the above assumptions:

• PRICE(settlement, maturity, rate, yld, redemption, frequency, [basis])=100.163338941426
• YIELD(settlement, maturity, rate, pr, redemption, frequency, [basis])=1.01125545674913%

Question:

• The YIELD function must return 1%, but it actually returns 1.01125545674913%.
• The difference 1% and 1.01125545674913% is too large if the difference is caused by using floating point numbers.
• It seems that the YIELD function doesn't work.  Is this a bug?

[#2]

@massanpc 

Perhaps that's http://westclintech.com/Blog/tabid/132/EntryId/74/Calculate-the-PRICE-of-a-bond-with-a-monthly-coupo... the reason. Jul31 and Sep30 are within one quarter, if take Jun30 result is more reliable

[#3]

@Sergei Baklan 

Thank you for the reply.

It makes sense.  On YIELD function - Office Support, the following remarks are posted:

• If there is one coupon period or less until redemption, YIELD is calculated as follows:
  

I've just calculated the above equation with the following numbers:

• par seems to be a typo in price/pr, and it is 100.163338941426
• A is equal to COUPDAYBS(settlement,maturity,frequency,[basis])=31
• E is equal to COUPDAYS(settlement,maturity,frequency,[basis])=90
• DSR=E-A=90-31=59

The YIELD function must return 0.0100000000000147891680, which is almost 1%.

I've also checked the calculation using Wolfram Alpha.

It seems that the YIELD function does not calculate as defined and the function is not reliable.