(L13)DataTilt

(L13)DataTilt - Topic #13 Data and Estimation Reading:...

Info iconThis preview shows pages 1–9. Sign up to view the full content.

View Full Document Right Arrow Icon
Primbs/Investment Science 1 Topic #13 Data and Estimation Reading: Luenberger Chapter 8, Sections 5 – 9
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Primbs/Investment Science 2 Data and Estimation Scaling Rules Effect of More Data Estimation of Parameters
Background image of page 2
Primbs/Investment Science 3 Data and Statistics We have been talking a lot about means and variances and covariances. In practice, we have to estimate these values. Let’s see how well we can estimate means and variances. ..
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Primbs/Investment Science 4 Mean and Variance Estimators = = n i i r n r 1 1 ˆ Mean: n r σ σ= ˆ Error: = - - = n i i r r n s 1 2 2 ) ˆ ( 1 1 Variance: 1 2 2 2 - = n s Note that the error in the mean estimate depends on the standard deviation. But, the error for the variance estimate is a fraction of the variance that is being estimated.
Background image of page 4
Primbs/Investment Science 5 Mean and Variance Estimators = = n i i r n r 1 1 ˆ Mean: = - - = n i i r r n s 1 2 2 ) ˆ ( 1 1 Variance: n r σ σ= ˆ Error: 1 2 2 2 - = n s Let’s consider estimating the monthly mean and variance for the market, and how many months we need for accurate estimates. # months 12 60 120 360 1200 mean error 1.25% 0.56% 0.40% 0.23% 0.13% variance error 8 x 10 -4 3.5 x 10 -4 2.4 x 10 -4 1.4 x 10 -4 7.6 x 10 -5
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Primbs/Investment Science 6 Mean and Variance Estimators = = n i i r n r 1 1 ˆ Mean: = - - = n i i r r n s 1 2 2 ) ˆ ( 1 1 Variance: n r σ σ= ˆ Error: 1 2 2 2 - = n s The mean is inherently more difficult to estimate than the variance. Using more data (daily for instance) will not improve your ability to estimate the mean. However, it will improve your ability to estimate the variance. Moral: (I) The mean return is difficult to estimate accurately. (II) The variance can be estimated will relative success.
Background image of page 6
Primbs/Investment Science 7 Covariance Estimators = - - - = n i i i r r r r r r n s 1 ) 2 ( ) 2 ( ) 1 ( ) 1 ( ) ˆ )( ˆ ( 1 1 2 1 To produce an unbiased estimate of the covariance use: Warning!: Excel divides by N instead of N-1 for its covariance estimator. This makes its covariance function inconsistent with its variance function!
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Primbs/Investment Science 8 Data and Estimation Scaling Rules Effect of More Data Estimation of Parameters
Background image of page 8
Image of page 9
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 28

(L13)DataTilt - Topic #13 Data and Estimation Reading:...

This preview shows document pages 1 - 9. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online