Thursday 29 January 2015, 1.00PM to 2:00pm
Speaker(s): Adriana Cornea-Madeira (York Management School)
We obtain the exact analytical expression for the weights in the Hodrick- Prescott (HP) filter. We then use the expression for the weights to build a fast algorithm with computational improvements by a factor of up to fifty times in samples typical in macroeconomics. Our expression for the weights gives insights about the properties of the HP filter and we use it to propose an end-point bias correction of the filter. We illustrate the bias correction on the estimation of the National Bureau of Economic Research (NBER) recession dates and find that our estimates are closer to the NBER dates when compared to the usual HP filter estimates. Finally, we show that our derivations for the weights provide a methodology for finding the exact weights of the more general Whittaker-Henderson filters of which the HP filter is a particular case.
Location: A/EC202 Economics Staff Room
Admission: Staff and PhD Students