HP 10bll+ Financial Calculator - How do i calculate sample covariance?

[Ajeff]

I'm studying for a business finance exam and need to know how to calculate the sample covariance in relation to securities on my calculator. I have entered the data necessary to calculate the mean return using the instructions from the

Mean, Standard Deviations, and Summation Statistics section of the manual. However i can't figure out how to use my calculator to quickly calculate the sample covariance. Any help or advice would be greatly appreciated. 🙂

[BartdB]

Let us take the following example:

http://www.wikihow.com/Calculate-Covariance

When all data is put into the calculator, you have MEAN(x), MEAN(y) and SUM(xy) provided by the calculator.

For the following (to help me type quicker) the "^" is blue upshift and "v" is orange downshift.

You can now calculate:

^ n - 1 = v 1/x * v ( ^ SUM(xy) - v ( ^ n * v MEAN(x,y) * v MEAN(x,y) v SWAP v ) v ) =

(you will actually get the answer -8.06944...  as n*mean(x)*mean(y) = 239.555' and not 239.41).

Key sequence as a picture:

[Ajeff]

That worked perfectly thank you :). I seem to have run into another problem however. I'm trying to work out the correlation coefficient between two stocks and i've been following the instructions in the manual and i keep getting error. The two stocks deviation on returns are 7.73% and 11.59% and the covariance of the stocks is -0.000656.

[BartdB]

Hi,

As demonstrated in my previous post, the 10BII+ does not know about covariance. It does some basic one & two variable statics.

So for the correlation coefficient it only knows how to determine this with actal data points for the two variables. However, in this case you just have the deviations and the covariance of the two variables.

So some other statistical identities are required:

HTH

.

[W0077815]

Seems complicated

Once pairs are entered

Why not workout the coefficient

Orange 4

Orange SWAP

Std Deviation for both

Orange 8

Orange Swap

Then the Covariance r=Coefficient*Std Dev x * Std Dev y

Cheers

[Maké]

HI!, W0077815:

Thank you by your participation, in the Forum, but this query, is very old (06-14-2013).

[BartdB]

---

@W0077815 wrote:

Seems complicated

Once pairs are entered

Why not workout the coefficient

Orange 4

Orange SWAP

Std Deviation for both

Orange 8

Orange Swap

Then the Covariance r=Coefficient*Std Dev x * Std Dev y

Cheers

---

You are right. That actually follows from my second post, thanks for pointing it out.