Auto correlation

[sahajiabhijit]

Analyst in his attempt to develop a model for predicting future correlation between two stock returns, perform auto regression to get below equation p(t) = 0.1 +0.3 p(t-1).

What will be the auto correlation for one period lag as calculated with respect to mean reversion?

a. 0.7
b. 0.10
c. 0.90
d. 0.30

 

[Nicole Seaman]

Analyst in his attempt to develop a model for predicting future correlation between two stock returns, perform auto regression to get below equation p(t) = 0.1 +0.3 p(t-1).

What will be the auto correlation for one period lag as calculated with respect to mean reversion?

a. 0.7
b. 0.10
c. 0.90
d. 0.30

Hello @sahajiabhijit

Can you please be specific about which GARP practice exam this question is from, along with the question number and explanation? I looked through the 2015, 2016 and 2017 GARP practice questions, and I do not see this question. It helps if we have all of this information so we do not have to search through a ton of documents to help answer. Since you've placed the question under the GARP practice section, I am assuming it is a GARP practice question?

Thank you,

Nicole

 

[sahajiabhijit]

this is not GARP question..

 

[Nicole Seaman]

this is not GARP question..

@sahajiabhijit

Thank you for confirming that this is not a GARP question. You posted it in the GARP practice question section, so I'm going to move it to a more appropriate section so other members are not confused or spend time trying to to look for it in one of the GARP practice exams. It would be helpful if you could post the complete answer and explanation also. It can be difficult for David and other members to answer questions like this when they don't know where it comes from, and they can't reference the source.

Nicole

 

[David Harper CFA FRM]

Hi @sahajiabhijit Per Diebold (or see also "Examples of low-order AR(p) processes for p = 1 at https://en.wikipedia.org/wiki/Autoregressive_model#Yule.E2.80.93Walker_equations ... well, the wikipedia shows the autocovariance but maybe doesn't solve through for the autocorrelation ... ) the correlation between observations that (h) periods apart is given simply by φ^h; or in your example, 0.30^1 = 0.30. I do not know what the phrase "with respect to mean reversion" means here. Thanks,