`bpass` (Series)

Band-pass filter

Syntax

[X, T] = bpass(X, Band, ~Range, ...)

Input arguments marked with a ~ sign may be omitted.

X [ tseries ]

Input tseries object that will be filtered.

Band [ numeric ]

Band of periodicities to be retained in the output data, Band = [Low, High].

~Range=Inf [ Dater ]

Date range on which the data will be filtered; if omitted, the entire time series range will be used.

X [ tseries ]

Band-pass filtered tseries object.

T [ tseries ]

Estimated trend tseries object.

AddTrend=true [ true | false ]

Add the estimated linear time trend back to filtered output series if band includes Inf.

Detrend=true [ true | false | cell ]

Remove an estimated time trend from the data before filtering; specify options for detrending in a cell array; see trend.

Log=false [ true | false ]

Logarithmize the data before filtering, de-logarithmize afterwards.

Method='cf' [ 'cf' | 'hwfsf' ]

Type of band-pass filter: Christiano-Fitzgerald, or h-windowed frequency-selective filter.

UnitRoot=true [ true | false ]

Assume unit root in the input data.

Christiano, L.J. and T.J.Fitzgerald (2003). The Band Pass Filter. International Economic Review, 44(2), 435--465.

Iacobucci, A. & A. Noullez (2005). A Frequency Selective Filter for Short-Length Time Series. Computational Economics, 25, 75--102.