Economics Insights @ CBS

In this paper, we measure the extent to which subjective and objective evaluations of new teachers in New York City can predict their future impacts on student achievement. Specifically, we examine evaluations of applicants to an alternative certification program, evaluations of new teachers by mentors that work with them during their first year, and evaluations based on student achievement data from their first year of teaching. We use a large sample, relative to prior work, and, unlike other studies (with the exception of John H. Tyler et al.

We consider an agent interacting with an unmodeled environment. At each time, the agent makes an observation, takes an action, and incurs a cost. Its actions can influence future observations and costs. The goal is to minimize the long-term average cost. We propose a novel algorithm, known as the active LZ algorithm, for optimal control based on ideas from the Lempel-Ziv scheme for universal data compression and prediction.

We establish that the min-sum message-passing algorithm and its asynchronous variants converge for a large class of unconstrained convex optimization problems, generalizing existing results for pairwise quadratic optimization problems. The main sufficient condition is that of scaled diagonal dominance. This condition is similar to known sufficient conditions for asynchronous convergence of other decentralized optimization algorithms, such as coordinate descent and gradient descent.

Social commerce is an emerging trend in which sellers are connected in online social networks, and where sellers are individuals instead of firms. This paper examines the economic value implications of a social network between sellers in a large online social commerce marketplace. In this marketplace each seller creates his or her own shop, and network ties between sellers are directed hyperlinks between their shops.