INTERPOL Procedure |
@INTERPOL is an old procedure for doing interpolation of a series to a higher frequency. The newer procedure for this is @DISAGGREGATE.
@INTERPOL(options) oldser newser
Parameters
|
oldser |
series to interpolate. This should be set up in the higher frequency (monthly in the example). @INTERPOL takes as the value to interpolate the last month of each quarter. |
|
newser |
higher frequency series to create |
Options
MODEL=[RW1]/AR1/RWAR1/RW2
Specifies one of several statistical models for the process. RW1 is a random walk, AR1 is a first order autoregression, RWAR1 is an ARIMA(1,1,0) and RW2 is ARIMA(0,2,0).
RHO=value of the AR1 parameter for AR1 and RWAR1 models [.9]
FACTOR=increase in the recording frequency (3 for quarterly to monthly)
Equivalent Use in @DISAGGREGATE
@INTERPOL(MODEL=model,RHO=rho,FACTOR=factor) seriesa seriesb
is equivalent to
@DISAGGREGATE(MAINTAIN=FINAL,TSMODEL=model,RHO=rho,FACTOR=factor) seriesa / seriesb
The RHO and FACTOR options have the same defaults in the two, so, in general, all you need to do to update a use of @INTERPOL to a newer @DISAGGREGATE is to add the MAINTAIN=FINAL option, replace MODEL with TSMODEL and add the / for the range between the two series.
Copyright © 2026 Thomas A. Doan