ARCH Garch Volatility

WHENWhen implementing data input for ARCH model initialization

ACTIONvalidate that input data contains only finite values using np.all(np.isfinite) before any numeric computation

CONSEQUENCEOptimizers and recursive variance computations will produce NaN/inf results, causing the entire model estimation to fail silently with meaningless outputs

WHENWhen implementing data input for ARCH model initialization

ACTIONconvert each input data to contiguous float64 arrays using np.ascontiguousarray before storing in self._y

CONSEQUENCENon-contiguous arrays or non-float64 types will cause buffer errors in Cython/Numba optimized recursive computations, leading to segmentation faults or incorrect variance calculations

WHENWhen initializing ARCH models with data

ACTIONpass None as input data without raising RuntimeError when attempting to fit the model

CONSEQUENCEFitting attempt with no data will cause cryptic errors in scipy optimize or segfault in Cython recursions