I searched for some web based papers and books, and read about cross-correlation for time series analysis. But I am trying to model a Multi-time series model. Do anyone know how can I do this?
You can use a vector auto-regressive model.... for example look at
https://www.researchgate.net/publication/258566784_Estimation_in_High-dimensional_Vector_Autoregressive_Models?ev=auth_pub
Article Estimation in High-dimensional Vector Autoregressive Models
So, is it possible to generate a formula like:
a*F(X,Y,Z)+bG(X,Z1,H)+cQ(V,Y,Z)+D=eU(X,T,Z)
with time series?
I mean replacing a time series for each X, Y etc and try to make an estimation model for U(X,T,Z).
I am not sure if this is of help, but to create an Auto Regressive model from a set of Time series, you can use the methods proposed by Gohberg in his book "Matrix Polynomials". Otherwise, feel free to take a look at the following conference papers, which include methods for consructing AR models that satisfy a given set of time series (i.e. discrete time functions).
Conference Paper On the Modeling of Discrete Time Auto-Regressive Representations
Conference Paper Modeling of Discrete Time Auto-Regressive Systems with given...
- Badges
- Science topic