You can estimate hidden markov models using the state space functions in S+FinMetrics 2.0. The state space functions allow Markov regime switching in the state space system matrices. Estimation, filt
For non Gaussian models, the Kalman filter is just the best linear forecasting model and is not the optimal filter. The likelihood estimation based on the normal distribution should be interpreted as
S+FinMetrics 2.0 contains a function gmm() for doing general linear and nonlinear GMM estimation. The function is very flexible. All you need to do is specify a function to compute the moment conditi
S+FinMetrics can handle a multivariate state space model. The specification of the measurement and transition equations allow for a multivariate response. I have some examples (see the term structur
One way to think of the irregular spacing is in terms of missing data. The true process works on equally spaced data but you ony observe data on specific dates. An AR type model can be fit using Kalm
The S+FinMetrics generic IC() will compute various information criteria (AIC, BIC, HQ and logLike) for various model fit objects. For example, IC.OLS() computes info criteria for OLS objects and IC.d
the book Analysis of Financial Time Series by Ruey Tsay (Wiley, 2002?) has a chapter on nonlinear modles including neural networks. For his examples he uses the nnet() function from the MASS library.
The problem is not in FinMetrics but in the S-PLUS function chol() and the type of Intel processor (P3, P4 etc). The function chol() is very sensitive to small departures from symmetric. The trick of
I believe that Splus 6.2 has a native interface to SQL server so that S-PLUS ODBC is no longer required. "Large" data set is all relative. Some people might think 1Mb is large, but that's actually ti
<?xml version="1.0" ?> Wolfgang You are not using the function aggregateSeries correctly. You should only be passing one time series (your ts). Your aggregation positions should be specified using a
The plot method for timeSeries objects, plot.timeSeries, works a bit differently than the other plot methods. To specify line types, colors etc you must use the plot.args() argument. For example, to
Dickey-Fuller tests are implemented in S+FinMetrics and the R package tseries. However, the DF unit root test is just a simple regression with lagged regressors and can be done using lm. You would ha
The SUR and NLSUR functions in Insightful's FinMetrics module can be used to fit linear and nonlinear seemingly unrelated regressions. Dear S-Plus users, I would like to perform a SUR-analysis in S-P
Sounds like you need to use a probabilistic discrete choice model (e.g. multinomial logit/probit or ordered logit/probit model). Insightful is developing a module for such analysis. Contact me if you
Ken Train has Gauss code for estimating a nested logit model using full information maximum likelihood on his UC Berkeley webpage (http://elsa.berkeley.edu/~train/ps.html). I believe the nested logit