Abstract:In Hotelling's model of spatial competition, a unit mass of voters is distributed in the interval $[0,1]$ (with their location corresponding to their political persuasion), and each of $m$ candidates selects as a strategy his distinct position in this interval. Each voter votes for the nearest candidate, and candidates choose their strategy to maximize their votes. It is known that if there are more than two candidates, equilibria may not exist in this model. It was unknown, however, how close to an equilibrium one could get. Our work studies approximate equilibria in this model, where a strategy profile is an (additive) $\epsilon$-equilibria if no candidate can increase their votes by $\epsilon$, and provides tight or nearly-tight bounds on the approximation $\epsilon$ achievable.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is how to quantify the quality of approximate equilibrium in Hotelling's spatial competition model when there are multiple candidates. Specifically:
1. **Problem Background**:
- In the Hotelling model, voters are distributed on the interval \([0, 1]\), and each candidate chooses a position within this interval as a strategy. Voters will vote for the candidate closest to them, and candidates maximize their number of votes by choosing positions.
- When the number of candidates exceeds two, there may be no exact equilibrium solution.
2. **Research Objectives**:
- This paper focuses on the state closest to equilibrium that can be achieved when there are three or more candidates. Specifically, the researchers defined the concept of an \(\epsilon\)-approximate equilibrium, that is, no candidate can increase their number of votes by more than \(\epsilon\) by changing their position.
- The main objective of the paper is to provide tight or nearly tight upper and lower bounds on \(\epsilon\).
3. **Main Contributions**:
- For the case of three candidates, the paper proves that for any voter distribution, \(\epsilon\geq\frac{1}{12}\). This means that no matter how the voters are distributed, there will always be at least \(\frac{1}{12}\) of the votes "left on the table", that is, they cannot be completely won by any one candidate.
- Further, the paper shows that in the worst - case scenario, there exists a voter distribution such that \(\epsilon\geq\frac{1}{6}\), and this bound is tight, that is, a \(\frac{1}{6}\)-approximate equilibrium can always be found.
- For the case of \(m\) candidates, the paper proves that as \(m\) increases, it is possible to get closer to the exact equilibrium. Specifically, for any voter distribution, a \(\frac{1}{m + 1}\)-approximate equilibrium can always be found in polynomial time, and this bound is asymptotically tight because there exists a voter distribution such that \(\epsilon\geq\frac{1}{m+3}\).
4. **Conclusions**:
- This research provides a quantitative perspective for understanding approximate equilibrium in the spatial competition model, especially when the number of candidates is large, the model becomes more stable.
- The research results also show that even in the worst - case scenario, a strategy configuration close to equilibrium can be found through an appropriate algorithm.
Through these studies, the paper fills the gap in the quantitative analysis of approximate equilibrium in the spatial competition model and provides an important theoretical basis for future research.