Professional survey forecasts and expectations in DSGE models

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F67985998%3A_____%2F23%3A00578223" target="_blank" >RIV/67985998:_____/23:00578223 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.cerge-ei.cz/pdf/wp/Wp766.pdf" target="_blank" >https://www.cerge-ei.cz/pdf/wp/Wp766.pdf</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Professional survey forecasts and expectations in DSGE models

Popis výsledku v původním jazyce

In this paper, we demonstrate the usefulness of survey data for macroeconomic analysis and propose a strategy to integrate and efficiently utilize information from surveys in the DSGE setup. We extend the set of observable variables to include the data on consumption, investment, output, and inflation expectations, as measured by the Survey of Professional Forecasters (SPF). By doing so, we aim to discipline the dynamics of model-based expectations and evaluate alternative belief models. Our approach to exploit the timely information from surveys is based on re-specification of structural shocks into persistent and transitory components. Due to the SPF, we are able to improve identification of fundamental shocks and predictive power of the model by separating the sources of low and high frequency volatility. Furthermore, we show that models with an imperfectly-rational expectation formation mechanism based on Adaptive Learning (AL) can reduce important limitations implied by the Rational Expectation (RE) hypothesis. More specifically, our models based on belief updating can better capture macroeconomic trend shifts and, as a result, achieve superior long-term predictions. In addition, the AL mechanism can produce realistic time variation in the transmission of shocks and perceived macro-economic volatility, which allows the model to better explain the investment dynamics. Finally, AL models, which relax the RE constraint of internal consistency between the agents’ and model forecasts, can reproduce the main features of agents’ predictions in line with SPF evidence and, at the same time, can generate improved model forecasts, thus diminishing possible inefficiencies present in surveys.

Název v anglickém jazyce

Professional survey forecasts and expectations in DSGE models

Popis výsledku anglicky

In this paper, we demonstrate the usefulness of survey data for macroeconomic analysis and propose a strategy to integrate and efficiently utilize information from surveys in the DSGE setup. We extend the set of observable variables to include the data on consumption, investment, output, and inflation expectations, as measured by the Survey of Professional Forecasters (SPF). By doing so, we aim to discipline the dynamics of model-based expectations and evaluate alternative belief models. Our approach to exploit the timely information from surveys is based on re-specification of structural shocks into persistent and transitory components. Due to the SPF, we are able to improve identification of fundamental shocks and predictive power of the model by separating the sources of low and high frequency volatility. Furthermore, we show that models with an imperfectly-rational expectation formation mechanism based on Adaptive Learning (AL) can reduce important limitations implied by the Rational Expectation (RE) hypothesis. More specifically, our models based on belief updating can better capture macroeconomic trend shifts and, as a result, achieve superior long-term predictions. In addition, the AL mechanism can produce realistic time variation in the transmission of shocks and perceived macro-economic volatility, which allows the model to better explain the investment dynamics. Finally, AL models, which relax the RE constraint of internal consistency between the agents’ and model forecasts, can reproduce the main features of agents’ predictions in line with SPF evidence and, at the same time, can generate improved model forecasts, thus diminishing possible inefficiencies present in surveys.

Druh

O - Ostatní výsledky

CEP obor

—

OECD FORD obor

50202 - Applied Economics, Econometrics

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2023

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů