chemrox
International Hazard
Posts: 2961
Registered: 18-1-2007
Location: UTM
Member Is Offline
Mood: LaGrangian
|
|
eignvectors
If you know how to apply linear algebra to geologic (or other) strain analysis please drop me u. Thanks, CRX
"When you let the dumbasses vote you end up with populism followed by autocracy and getting back is a bitch." Plato (sort of)
|
|
|
battoussai114
Hazard to Others
Posts: 235
Registered: 18-2-2015
Member Is Offline
Mood: Not bad.... Not bad.
|
|
I know a bit about application of linear algebra to DEs systems... Which are used in solid mechanics (though I don't know much about the later).
Batoussai.
|
|
|
AJKOER
Radically Dubious
Posts: 3026
Registered: 7-5-2011
Member Is Offline
Mood: No Mood
|
|
If your interest relates to probabilistic change in sediment structures with time, the particular area of interest could be addressed by Markov
Transition Matrices. Eigenvectors can be used to transform matrices to diagonal form. The derived diagonal elements are the so called eigenvalues.
Taking these values to various powers and re-applying the eigenvector transformation is a path to constructing transition matrices for various time
frames.
Otherwise, the apparent application of linear algebra, a back bone of statistical theory, to strain analysis may fall in the realm of failure
analysis, failure distributions, ... discussed in courses on Reliability Theory.
Predicting when a bank will fail is not too far a field from the likelihood of a fault line erupting. Matrices of correlated variables producing a
discriminate function (field is Discriminate Analysis) to predict a binary event (yes or no), with a preselected Type I and Type ll Error.
What statistics does not cover is the art of selecting important variables apart from correlations (based on ones knowledge of the field) or
constructing indicator variables (imagination as to the underlying model, in the business world, for example, this may be based on economic theory) to
arrive at good prediction models. Generally, too many variables produce poor prediction models.
[Edit] While I have some background in the above areas, I have not made much ground in the field of Spatial Statistics, which may also have some
application to geological features. However, I would not recommend entering this realm until you have a good grasp of what I alluded to above.
[Edited on 29-8-2016 by AJKOER]
|
|
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
