Date of Award




Document Type


Degree Name

Doctor of Philosophy (PhD)


Department of Economics

Content Description

1 online resource (ii, vi, 114 pages) : illustrations (some color)

Dissertation/Thesis Chair

John B Jones

Committee Members

Betty C Daniel


dynamic stochastic general equilibrium model, entry and exit of firms, uncertainty, Uncertainty, Risk, Economic forecasting, Employment forecasting, Forecasting, Measurement uncertainty (Statistics)

Subject Categories



In this dissertation, I study how uncertainty interacts with firm dynamics to affect aggregate economic variables. The first chapter proposes a measure of uncertainty based on forecaster surveys and compares it with another measure developed by Jurado, Ludvigson and Ng (2015). This chapter first investigates whether fluctuations in survey composition make the survey uncertainty measure too dependent on the identity of participating forecasters. An extension of the Breusch and Pagan (1979) test due to Breitung, Roling and Salish (2016) rejects constancy of parameters in regressions of GDP growth on survey uncertainty as the subsets of the forecasters differ. However, the predictions for GDP growth from the regressions do not vary much depending on which subset of the forecasters is used, suggesting that the survey uncertainty measure does not depend so much on composition as to make it uninformative. Next I compare the survey uncertainty measure with the financial uncertainty measure constructed by Ludvigson, Ma and Ng (2016) based on the method of Jurado, Ludvigson and Ng (2015), in terms of their relationships with GDP growth. The relationship between GDP growth and the financial uncertainty measure is found to have two structural breaks over the period 1980-2016. When the sample period is divided into three sub-periods delimited by the break dates, the coefficients on financial uncertainty in the GDP growth equation of a VAR change dramatically across periods. The change in the coefficients on survey uncertainty are less pronounced, possibly because it has a weaker association with GDP growth in general. Because of this weaker association of the survey uncertainty measure with growth, I use Ludvigson, Ma and Ng's (2016) measure of financial uncertainty as my preferred measure in the second chapter.

Included in

Economics Commons