Search

Objective: We propose a simple approach to dealing with partial contestation in models of multiparty elections. Methods: Our proposed approach is to add a tiny value to the vote share of parties that do not contest a district and then to include dummy variables identifying those districts in which parties do not compete. We can then estimate a single system of equations using a seemingly unrelated regression (SUR) approach and Aitchison's log‐ratio transformation. In our SUR system, we interact the dummy variable for a party that partially contested districts with other predictors in the equation that uses the share of votes of the same party in the log‐ratio outcome. Finally, we estimate robust standard errors for predictors in this equation to address heteroscedasticity. Results: We demonstrate the utility of our approach using simulated data and election results from the English parliamentary elections in 2017. Conclusion: From our simulations, we find that our recommended approach performs as well as that proposed by Tom, Tucker, and Wittenberg. Our strategy is advantageous in that it is easy to estimate, uses information from all districts, and addresses partial contestation in real‐world elections with a single system of seemingly unrelated regressions. [ABSTRACT FROM AUTHOR]

Objective: The study aims to demonstrate the utility of modeling compositional volatility in substantive domains beyond budgeting. Methods: We show how to model compositional volatility on its own or as a part of a system of equations in which the component parts of the compositional outcome variable are also modeled. Results: Using data on the volatility of support for German political parties, we demonstrate the usefulness of stand‐alone models of compositional volatility. Using data on the volatility of income shares in the United States, we demonstrate the usefulness of modeling volatility together with compositional components. Conclusion: There is considerable potential for modeling compositional volatility. [ABSTRACT FROM AUTHOR]

Monte Carlo simulations are commonly used to test the performance of estimators and models from rival methods, under a range of data-generating processes. This tool improves our understanding of the relative merits of rival methods in different contexts, such as varying sample sizes and violations of assumptions. When used, it is common to report the bias or the root mean squared error of the different methods. It is far less common to report the standard deviation, overconfidence, coverage probability, or power. Each of these six performance statistics provides important, and often differing, information regarding a method's performance. Here, we present a structured way to think about Monte Carlo performance statistics. In replications of three prominent papers, we demonstrate the utility of our approach and provide new substantive results about the performance of rival methods. [ABSTRACT FROM AUTHOR]