|
on Discrete Choice Models |
| By: | Mark J. Koetse (Vrije Universiteit Amsterdam); Henri L.F. de Groot (Vrije Universiteit Amsterdam); Raymond J.G.M. Florax (Purdue University, and Vrije Universiteit) |
| Abstract: | In this paper we perform a meta-analysis on empirical estimates of the impact between investment and uncertainty. Since the outcomes of primary studies are largely incomparable with respect to the magnitude of the effect, our analysis focuses on the direction and statistical significance of the relationship. The standard approach in this situation is to estimate an ordered probit model on a categorical estimate, defined in terms of the direction of the effect. The estimates are transformed into marginal effects, in order to represent the changes in the probability of finding a negative significant, insignificant, and positive significant estimate. Although a meta-analysis generally does not allow for inferences on the correctness of model specifications in primary studies, our results give clear directions for model building in empirical investment research. For example, not including factor prices in investment models may seriously affect the model outco! mes. Furthermore, we find that Q-models produce more negative significant estimates than other models do, ceteris paribus. The outcome of a study is also affected by the type of data used in a primary study. Although it is clear that meta-analysis cannot always give decisive insights into the explanations for the variation in empirical outcomes, our meta-analysis shows that we can explain to a large extent why empirical estimates of the investment-uncertainty relationship differ. |
| Keywords: | investment; uncertainty; meta-analysis |
| JEL: | C10 D21 D80 E22 |
| Date: | 2006–07–10 |
| URL: | http://d.repec.org/n?u=RePEc:dgr:uvatin:20060060&r=dcm |
| By: | Gert-Jan M. Linders (Faculty of Economics and Business Administration, Vrije Universiteit Amsterdam); Henri L.F. de Groot (Faculty of Economics and Business Administration, Vrije Universiteit Amsterdam) |
| Abstract: | The gravity model is the workhorse model to describe and explain variation in bilateral trade patterns. Consistent with both Heckscher-Ohlin models and models of imperfect competition and trade, this versatile model has proven to be very successful, explaining a large part of the variance in trade flows. However, the log-linear model cannot straightforwardly account for the occurrence of zero-valued trade flows between pairs of countries. This paper investigates the various approaches suggested to deal with zero flows. Apart from the option to omit the zero flows from the sample, various extensions of Tobit estimation, truncated regression, probit regression and substitutions for zero flows have been suggested. We argue that the choice of method should be based on both economic and econometric considerations. The sample selection model appears to fit both considerations best. Moreover, we show that the choice of method may matter greatly for the results. In the end, the results surprisingly suggest that the simplest solution, to omit zero flows from the sample, often leads to acceptable results, although the sample selection model is preferred theoretically and econometrically. |
| Keywords: | bilateral trade; gravity model; zero flows; sample selection model |
| JEL: | F14 F15 |
| Date: | 2006–08–14 |
| URL: | http://d.repec.org/n?u=RePEc:dgr:uvatin:20060072&r=dcm |