So far, we saw that, if the OMT is rejected, there is a problem in one of the three components of the mathematical model: observations, the functional model, or in the stochastic model. In this video, we will look at a line fitting example, nd elaborate more on these three sources of the rejection.

Note: below the video you can find a small text unit with a correction.

Correction to the explanation in the video:

In this video at 4’51” , it was mentioned that we had used a new =5.3 as the new variance instead of the initial variance of =2. However, the correct procedure (which has been applied in the shown example), is to scale-up the initial variance by the factor 5.3. That is, we used a new variance of 5.3×2=10.6 in order to adapt the stochastic model. (See the summary of this section, for more information why we scale-up the covariance matrix by the factor