Search

International audience; Skewness of return has been suggested as a reason why agents might choose to gamble, ceteris paribus, in Cumulative Prospect Theory (CPT). We investigate the relationship between moments of return in two models where agents choices over uncertain outcomes are determined as in CPT. We illustrate via examples that in CPT theory, as with expected utility theory, propositions that agents have a preference for skewness may be invalid.

International audience; Standard parametric specifications of Cumulative Prospect theory (CPT) can explain why agents bet on longshots at actuarially unfair odds. However the standard specification of CPT cannot explain why people might bet on more favored outcomes, where by construction the greatest volume of money is bet. This paper outlines a parametric specification than can consistently explain gambling over all outcomes. In particular we assume that the value function is bounded from above and below and that the degree of loss aversion experienced by the agent is smaller for small-stake gambles (as a proportion of wealth) than usually assumed in CPT. There are a number of new implications of this specification. Boundedness of the value function in CPT implies that the indifference curve between expected-return and win-probability for a given stake will typically exhibit both an asymptote (implying rejection of an infinite gain bet) and a minimum, as the shape of the value function dominates the probability weighting function. Also the high probability section of the indifference curve will exhibit a maximum.

Recent research has reported the lack of correct size in stationarity test for PPP deviations within a linear framework. However, theoretically well motivated nonlinear models, such as the ESTAR, appear to parsimoniously fit the PPP data and provide an explanation for the PPP ?puzzle?. Employing Monte Carlo experiments we analyze the size and power of the nonlinear tests against a variety of nonstationary hypotheses. We also fit the ESTAR model to data from high inflation economies. Our results provide further support for ESTAR specification.

In two recent contributions Lothian and Taylor, and Cuddington and Liang, produced empirical evidence that annual data for the dollar-sterling real exchange rate spanning two centuries exhibited a non-linear deterministic trend. This trend could be proxying Harrod-Balassa-Samuelson effects. Lothian and Taylor showed that a linear stationary autoregressive mode, which embodied a cubic trend, implied much faster mean reversion of the real exchange rate to shocks than a model that excluded the trend. This article shows that both non-linearity and a deterministic trend can be allowed for in a theoretically appealing manner and that the fitted models provide a parsimonious explanation of both the dollar-sterling and franc-sterling real exchange rates over the two centuries of data. Generalized impulse response function analysis of the models demonstrates that the speed of adjustment to shocks can be even faster when trends are considered.

Bookmakers practise a type of product bundling. To bet a horse for a place a punter has to bet an equal amount for a win. The returns to the place component of the bet are determined by a rule of thumb. This paper examines whether the product bundling negates a betting strategy that endeavours to exploit any inefficiency in pricing of the place component of each way bets based on the Harville formulae. In order to implement the Harville formulae Shin probabilities are employed, which correct for the favourite longshot bias in starting price odds, as measures of win probabilities. The analysis suggests that small positive expected returns to each way bets appear to exist.

We employ the Brock, Dechert and Scheinkman statistic to investigate non-linearity in conditional mean in UK real wages, employment and output, and to fit threshould regression models to the series. The latter porvide reasonable representations for thenon-lineraity in real wages and output but the residuals from the threshold autoregressive model for employment still contain a non-linear component. In a forecast comparison the TAR models performed better than simple autoregressions. Using the lagged share of wages as an error-correction term we find evidence that all three series respond asymmetrically to lagged changes in wage share and other variables. Fore-casts of real wages and employment derived from these models were superior to those derived from standard, symmetric, ECMs.

(1975). The determinants of unemployment and the ‘UV’ relationship. Applied Economics: Vol. 7, No. 4, pp. 251-255.

(1979). Price expectations: workers versus capitalists. Applied Economics: Vol. 11, No. 3, pp. 235-240.

This paper examines how time series models of gross domestic product and its components in the UK change as a result of both new observations becoming available and the provisional data series being revised. Quarterly predictions from the time series models are compared with the series of first estimates of each quarterly change, the series available one year after the forecasts are made and the most recent data available. The time series models are found to be satisfactory in a number of cases and are better predictors of the final values of total final expenditure than the first estimates series. Surprisingly, the first estimates series are closer to the most recent data available than the estimates obtained one year after the data period ended.