I discuss models which allow the local level model, which rationalised exponentially weighted moving averages, to have a time-varying signal/noise ratio.  I call this a martingale component model.  This makes the rate of discounting of data local.  I show how to handle such models...

Estimating the covariance and correlation between assets using high frequency data is challenging due to market microstructure effects and Epps effects.  In this paper we extend Xiu's univariate QML approach to the multivariate case, carrying out inference as if the observations arise from an...

High frequency financial data allows us to learn more about volatility, volatility of volatility and jumps.  One of the key techniques developed in the literature in recent years has been bipower variation and its multipower extension, which estimates time-varying volatility robustly to...

We consider an observer who makes a finite number of observations of an industry producing a homogeneous good, where each observation consists of the market price and firm specific production quantities.  We develop a revealed preference test (in the form of a linear program) for the hypothesis...

This paper discusses the incentive to bundle when consumer valuations are non-additive and/or when products are supplied by separate sellers.  Whether integrated or separate, a firm has an incentive to introduce a bundle discount when demand for the bundle is more elastic than the overall...

We study price competition between firms over public list or posted prices when a fraction of consumers (termed 'bargainers') can subsequently receive discounts with some probability.  Such stochastic discounts are a feature of markets in which some consumers bargain explicitly; of markets in...

This paper examines the operation of an emissions trading scheme (ETS) in a Cournot oligopoly.  We study the impact of the ETS on industry output, price, costs, emissions, and profits.  In particular, we develop formulae for the number of emissions permits that have to be freely allocated to...

We study two platforms competing for members by investing in network quality.  Quality is complementary to the network size: the marginal utility generated by an additional member increases with the network's quality.  Platforms are imperfect substitutes: a share of the potential members are...

Search engines enable advertisers to target consumers based on the query they have entered.  In a framework with horizontal product differentiation, imperfect product information and in which consumers incur search costs, I study a game in which advertisers have to choose a price and a set of...

This paper surveys models of markets in which some consumers are "savvy" while others are not.  We discuss when the presence of savvy consumers improves the deals available to non-savvy consumers in the market (the case of search externalities), and when the non-savvy fund generous deals for...