The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.

Model-selection-MRHLP

Introduction

In this package, it is possible to select models based on information criteria such as BIC, AIC and ICL.

The selection is done on two parameters which are:

\(K\): The number of regimes;
\(p\): The order of the polyniomial regression.

Data

Let’s select a MRHLP model for the following multivariate time series \(Y\):

data("multivtoydataset")
x <- multivtoydataset$x
y <- multivtoydataset[, c("y1", "y2", "y3")]
matplot(x, y, type = "l", xlab = "x", ylab = "Y", lty = 1)

Model selection with BIC

selectedmrhlp <- selectMRHLP(X = x, Y = y, Kmin = 2, Kmax = 6, pmin = 0, pmax = 3)
## The MRHLP model selected via the "BIC" has K = 5 regimes 
##  and the order of the polynomial regression is p = 0.
## BIC = -3033.20042397111
## AIC = -2913.75756459291

The selected model has \(K = 5\) regimes and the order of the polynomial regression is \(p = 0\). According to the way \(Y\) has been generated, these parameters are what we expected.

Let’s summarize the selected model:

selectedmrhlp$summary()
## ----------------------
## Fitted MRHLP model
## ----------------------
## 
## MRHLP model with K = 5 regimes
## 
##  log-likelihood nu       AIC     BIC       ICL
##       -2860.758 53 -2913.758 -3033.2 -3032.414
## 
## Clustering table:
##   1   2   3   4   5 
## 100 120 200 100 150 
## 
## 
## ------------------
## Regime 1 (K = 1):
## 
## Regression coefficients:
## 
##   Beta(d = 1) Beta(d = 2) Beta(d = 3)
## 1   0.1131005   0.9124035   -1.850509
## 
## Covariance matrix:
##                                    
##  1.19064699  0.12700417  0.05496662
##  0.12700417  0.90279499 -0.03272115
##  0.05496662 -0.03272115  0.89086804
## ------------------
## Regime 2 (K = 2):
## 
## Regression coefficients:
## 
##   Beta(d = 1) Beta(d = 2) Beta(d = 3)
## 1    7.190579    5.049538    9.952361
## 
## Covariance matrix:
##                                   
##   1.0723960 -0.18151782 0.12179798
##  -0.1815178  1.05340358 0.01211349
##   0.1217980  0.01211349 0.76527294
## ------------------
## Regime 3 (K = 3):
## 
## Regression coefficients:
## 
##   Beta(d = 1) Beta(d = 2) Beta(d = 3)
## 1    3.951224    5.941976    7.950232
## 
## Covariance matrix:
##                                     
##   1.02880640 -0.05856588 -0.02543545
##  -0.05856588  1.19527262  0.02309638
##  -0.02543545  0.02309638  1.01201958
## ------------------
## Regime 4 (K = 4):
## 
## Regression coefficients:
## 
##   Beta(d = 1) Beta(d = 2) Beta(d = 3)
## 1  -0.9461282   -1.901665   0.0135667
## 
## Covariance matrix:
##                                     
##   0.88092255 -0.02771294 -0.03959332
##  -0.02771294  1.14567525 -0.10726848
##  -0.03959332 -0.10726848  0.89325063
## ------------------
## Regime 5 (K = 5):
## 
## Regression coefficients:
## 
##   Beta(d = 1) Beta(d = 2) Beta(d = 3)
## 1    3.549319    1.888239    4.981038
## 
## Covariance matrix:
##                                 
##  1.1333159 0.25852405 0.03091090
##  0.2585241 1.21453178 0.05663565
##  0.0309109 0.05663565 0.84026581

selectedmrhlp$plot(what = "estimatedsignal")

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.