We show that technical indicators deliver stable economic value in predicting the U.S. equity premium over the out-of-sample period from 1966 to 2014. Results tentatively improve over time and beat alternatives over a large continuum of sub-periods. By contrast, economic indicators work well...

We develop and implement a new measure for inequality aversion: two peers are endowed with identical binary lotteries and the only choice they make is whether they want to play out the lotteries independently or with perfect positive correlation (coupling). Coupling has no other e ect than...