Forecasting Inflation Rates Using Daily Data: A Nonparametric MIDAS Approach

Forecasting Inflation Rates Using Daily Data: A Nonparametric MIDAS Approach

0.00 Avg rating0 Votes
Article ID: iaor201528907
Volume: 34
Issue: 7
Start Page Number: 588
End Page Number: 603
Publication Date: Nov 2015
Journal: Journal of Forecasting
Authors: ,
Keywords: simulation: applications
Abstract:

In this paper a nonparametric approach for estimating mixed‐frequency forecast equations is proposed. In contrast to the popular MIDAS approach that employs an (exponential) Almon or Beta lag distribution, we adopt a penalized least‐squares estimator that imposes some degree of smoothness to the lag distribution. This estimator is related to nonparametric estimation procedures based on cubic splines and resembles the popular Hodrick–Prescott filtering technique for estimating a smooth trend function. Monte Carlo experiments suggest that the nonparametric estimator may provide more reliable and flexible approximations to the actual lag distribution than the conventional parametric MIDAS approach based on exponential lag polynomials. Parametric and nonparametric methods are applied to assess the predictive power of various daily indicators for forecasting monthly inflation rates. It turns out that the commodity price index is a useful predictor for inflations rates 20–30 days ahead with a hump‐shaped lag distribution.

Reviews

Required fields are marked *. Your email address will not be published.