The major U.S. stock indexes were mostly lower in morning trading Friday as investors sized up the latest batch of company earnings. Real estate stocks lagged the most, while energy companies led the gainers as the price of crude oil headed higher. Several technology companies were moving higher after reporting solid results.
British Chancellor of the Exchequer George Osborne speaks during a news conference at The Treasury in London, Monday, June 27, 2016. Treasury chief Osborne sought to calm nerves in the markets Monday, as investors worried about the consequences of Britain leaving the European Union. In his first public appearance since the vote to leave the bloc Thursday, Osborne tried to reassure markets shaken by the result.
The coincident index of economic indicators fell for the first time in three months in August, the government said in a preliminary report Friday. The coincident index, which reflects the current state of the economy based on industrial output, retail sales, new job offers and other factors, fell 0.1 point from the previous month to 112.0, against the 2010 base of 100, the Cabinet Office said. The leading index, which predicts developments in the coming few months, rose 1.2 points to 101.2, while the lagging index, which measures economic performance in the recent past, rose 0.3 point to 113.4. The government maintained its basic assessment of the coincident index as "pausing."
Starting with the Thomspon sampling algorithm, recent years have seen a resurgence of interest in Bayesian algorithms for the Multi-armed Bandit (MAB) problem. These algorithms seek to exploit prior information on arm biases and while several have been shown to be regret optimal, their design has not emerged from a principled approach. In contrast, if one cared about Bayesian regret discounted over an infinite horizon at a fixed, pre-specified rate, the celebrated Gittins index theorem offers an optimal algorithm. Unfortunately, the Gittins analysis does not appear to carry over to minimizing Bayesian regret over all sufficiently large horizons and computing a Gittins index is onerous relative to essentially any incumbent index scheme for the Bayesian MAB problem. The present paper proposes a sequence of'optimistic' approximations to the Gittins index.