gLog.ep: Adjust for zero values and compute log[abs(e)^p]
Description
Adjusts a series called e - typically a series of residuals - for zero values,
so that the logarithm can be applied on the absolute pth. exponentiated
values. Next, log[abs(e)^p] is computed
Usage
gLog.ep(e, zero.adj=0.1, p=2, na.replace=NA)
Arguments
e
numeric vector, time series or zoo object
zero.adj
numeric value between 0 and 1. The quantile adjustment for zero values.
The default 0.1 means that the zero residuals are replaced by means of
the 10 percent quantile of the absolute residuals before taking the
logarithm
p
numeric value greater than zero. The power of the log-volatility
specification.
na.replace
the value to replace NA values with. Default: na.replace=NA
Value
log[abs(e)^p], where the e values have been zero adjusted
References
Genaro Sucarrat and Alvaro Escribano: 'Automated Financial Model Selection:
General-to-Specific Modelling of the Mean and Volatility Specifications',
forthcoming in the Oxford Bulletin of Economics and Statistics