Evaluating Forecasts, Narratives and Policy Using a Test of Invariance

Oxford Review of Economic Policy

Muellbauer

2018

The adoption as policy models by central banks of representative agent New Keynesian dynamic stochastic general equilibrium models has been widely criticised, including for their simplistic micro-foundations. At the Bank of England, the previous generation of policy models is seen in its 1999 medium-term macro model (MTMM). Instead of improving that model to correct its considerable flaws, many shared by other non-DSGE policy models such as the Federal Reserve's FRB/US, it was replaced in 2004 by the DSGE-based BEQM. Though this clearly failed during and after the global financial crisis, it was replaced in 2011 by the DSGE COMPASS, complemented by a 'suite of models'. We provide a general critique of DSGE models for explaining, forecasting and policy analyses at central banks, and suggest new directions for improving current empirical macroeconomic models based on empirical modelling broadly consistent with better theory, rather than seeking to impose simplistic and unrealistic theory.

Section: (Ii) Forecasting Failuresmentioning

confidence: 99%

The future of macroeconomics: macro theory and models at the Bank of England

Oxford Review of Economic Policy

Muellbauer

2018

“…The correlation is effectively zero over 2020. To disentangle this effect, we apply a subset of multiplicative indicator saturation (MIS; see Castle et al, 2017) to identify parameter nonconstancy in the model. We include π t Â S 2020 j ð Þ for j ¼ 1,…, 9, where S 2020 j ð Þ is a step indicator that takes the value 1 for observations 2020(1)-2020(j) [2020(9) is the end of the sample], in model (2).…”

Section: Forecasting the Uk Unemployment Rate Over The Pandemicmentioning

confidence: 99%

The Value of Robust Statistical Forecasts in the Covid-19 Pandemic

Castle

Doornik

2021

NIER

The Covid-19 pandemic has put forecasting under the spotlight, pitting epidemiological models against extrapolative time-series devices. We have been producing real-time short-term forecasts of confirmed cases and deaths using robust statistical models since 20 March 2020. The forecasts are adaptive to abrupt structural change, a major feature of the pandemic data due to data measurement errors, definitional and testing changes, policy interventions, technological advances and rapidly changing trends. The pandemic has also led to abrupt structural change in macroeconomic outcomes. Using the same methods, we forecast aggregate UK unemployment over the pandemic. The forecasts rapidly adapt to the employment policies implemented when the UK entered the first lockdown. The difference between our statistical and theory based forecasts provides a measure of the effect of furlough policies on stabilising unemployment, establishing useful scenarios had furlough policies not been implemented.

“…Kitov and Tabor (2015) have investigated the properties of MIS by simulation, and found it can detect shifts in regression parameters despite the huge number of candidate variables. This prompted Castle et al (2017) to apply the approach to successfully detect induced shifts in estimated models following a policy intervention. They offer an explanation for the surprisingly good performance of MIS as follows.…”

Section: Multiplicative-indicator Saturationmentioning

confidence: 99%

Detectives of Change: Indicator Saturation

Castle

Modelling Our Changing World

2019

Structural changes are pervasive from innovations affecting many disciplines. These can shift distributions, altering relationships and causing forecast failure. Many empirical models also have outliers: both can distort inference. When the dates of shifts are not known, they need to be detected to be handled, usually by creating an indicator variable that matches the event. The basic example is an impulse indicator equal to unity for the date of an outlier and zero elsewhere. We discuss an approach to finding multiple outliers and shifts called saturation estimation. For finding outliers, an impulse indicator is created for every observation and the computer program searches to see which, if any, match an outlier. Similarly for location shifts: a step indicator equal to unity till time t is created for every t and searched over. We explain how and why this approach works.