On Thu, Oct 18, 2012 at 1:37 AM, Didier Vila <[email protected]> wrote:
> Problem 1: My problem is a risk problem at the moment.
>
> ** **
>
> **· **I want to represent the behaviour of my 20 000 time series
> and generates some monte carlo simulations using the method.sample()****
>
> **· **Intuitively, i have to choose between 1 or 3 ( max 4) the
> number of states.****
>
> **· **Then, I want to capture the risk for each time series and
> the risk at the aggregate level.( I will generate 100 Monte Carlo
> Simulations)
>
Alright, then optimizing "score" may make sense it that case.
> ****
>
> **· **Question 1: I don t see how I can train and cross validate
> my HMM in Scikit Learn ( First time I use Scikit Learn for this purpose)*
> ***
>
> **· **Question 2: the lenght of the time serie is 32 periods, is
> it enough to make cross validation and validation ?
>
Have a look at http://scikit-learn.org/0.12/modules/grid_search.html.
> ****
>
> ** **
>
> Problem 2: Classification****
>
> **· **In the near future, I will try to make some classification
> of time series but I have no ideas how to handle the problem ? Should I use
> an SVM ? Can you refer any paper ?
>
You can group your time series per class and train one HMM per class with
those time series. Then given a new time series, you can decide its class
by the argmax of the Bayes rule:
Class = argmax P(Class | Time Series) = argmax P(Time Series | Class) *
P(Class) / P(Time Series) = argmax P(Time Series | Class) * P(Class)
P(Time Series | Class) can be computed by the forward algorithm or can be
approximated by the Viterbi algorithm (which is more numerically stable).
P(Class) can be computed by counting the number of time series in each
class.
> Generic Questions: I was wondering if your algorithm is developed iin
> Python ? Do you think your algo is relevant to apply to my business problem
> ?
>
My method is useful for classifying time series which are made of smaller
parts whose label you don't know. So I don't think it would work for you.
Mathieu
------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_sfd2d_oct
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
