isChi leads to R to crash

[szhan]

When running runCSSR() with isChi set to TRUE, R encounters a fatal error. The error seems to go away when time series is long enough (say, >30).

When running CSSR with the flag -ch, the following error (and related ones) was encountered:
"a too large, ITMAX too small in continued fraction gamma function"

I suspect that the crash may be avoided if nerror() in Test.cpp is handled better within the Rcpp framework. For now, it is best to avoid performing chi-square test altogether (i.e., leaving isChi=FALSE).

[szhan]

This is probably related to one of the known issues of the C++ source code...