Two Trees

We solve a model with two i.i.d. Lucas trees. Although the correspondingone-tree model produces a constant price-dividend ratio andi.i.d. returns, the two-tree model produces interesting asset-pricingdynamics. Investors want to rebalance their portfolios afterany change in value. Because the size of the trees is fixed,prices must adjust to offset this desire. As a result, expectedreturns, excess returns, and return volatility all vary throughtime. Returns display serial correlation and are predictablefrom price-dividend ratios. Return volatility differs from cash-flowvolatility, and return shocks can occur without news about cashflows.     

Explaining economic growth in the Soviet Union, 1950–86

Production function estimates are provided for Soviet industrial production and gross national product for the period 1950–86. A variety of alternative specifications is tested, including Cobb-Douglas, constant elasticity of substitution and variable elasticity of substitution production functions, and an error correction mechanism is used to investigate the long-run properties of the estimated equation. The structural stability of the estimates is also examined. Constant-returns-to-scale Cobb-Douglas production functions suggest that the rate of total factor productivity growth in the Soviet economy has declined steadily over time, becoming negative sometime in the period between 1970 and 1980. However the extensive statistical tests can doubt on the validity of any production function estimated on Soviet data.     

The Opening, Short, Medium and Long Term Performance of Canadian Unit Trust Initial Public Offerings (IPOs)

Abstract:   This paper investigates the initial pricing and performance of Canadian unit trust IPOs over a three‐ to four‐year period and then draws implications for the efficiency of the Canadian market. Overall, the results confirm the following: in the short term, unit trust IPOs are underpriced and outperform the Canadian market; in the medium term, IPOs are fairly priced and neither outperform nor underperform the Canadian market; and in the long term, IPOs are fairly priced but underperform the Canadian market. In addition, our results confirm that the size of underpricing is related to ex‐ante uncertainty about the value of the issue. Ex‐ante uncertainty proxies, namely total risk, exchange listing, relative bid‐ask spread, and relative volume of initial trade, all explain the size of underpricing. When the effects of these factors are controlled, the results confirm that Canadian unit trust IPOs are indeed overpriced in the short term but underpriced in the long term. We conclude that the Canadian unit trust IPO market appears to be inefficient in the short and long term, but over the medium, the market appears to be efficient.     

A Cox Parametric Bootstrap Test of the von Liebig Hypotheses

This study uses a Cox parametric bootstrap test to select between two specifications of the von Liebig hypothesis, a switching regression (SR) model, and a linear response function with a stochastic plateau. Specifying the production function as a linear response function with a stochastic plateau yields a superior approximation of the data for livestock gain as a function of forage allowance than the SR approach.
Dans la présente étude, nous avons utilisé la technique du bootstrap paramétrique de Cox pour choisir entre deux caractéristiques de l'hypothèse de von Liebig: un modèle de régression avec changement de régime et une fonction de réponse linéaire avec plateau stochastique. établir la fonction de production comme une fonction de réponse linéaire avec plateau stochastique offre une meilleure approximation des données sur le gain de poids du bétail comme fonction de l'apport de fourrages comparativement au modèle de régression avec changement de régime.     

Long-run Relationships Between World Vegetable Oil Prices

In the international edible oil markets, there is believed to be high substitutability between vegetable oils and fats produced under different conditions. In light of this, we consider the question: what is the nature of the long-run relationships between vegetable oil prices? Long-run co-movements among oil prices are analysed, based on a multivariate cointegration model. The empirical finding is that most co-movements are consistent with the predictions of market theory. Prices of oils tend to be grouped according to their different end-uses. Some policy implications of a buffer stock scheme are discussed.     

Analysis of efficient markets

A simple trading model is presented in which Bayes’ rule is used to aggregate traders’ forecasts about risky assets’ future returns. In this financial market, Bayes’ rule operates like an omnipotent market-maker performing functions that in 1776 Adam Smith attributed to an “invisible hand.” We have analyzed two distinct cases: in the first scenario, the traders’ forecast errors are uncorrelated, and in the second scenario, the traders’ forecast errors are correlated. The contribution of our paper is fourfold: first, we prove that the “efficient market” mean-return can be expressed as a complex linear combination of the traders’ forecasts. The weights depend on the forecast variances, as well as on the correlations among the traders’ forecasts. Second we show that the “efficient” variance is equal to the inverse of the sum of the traders’ precision errors, and is also related to the correlations among the traders’ forecast errors. Third, we prove that the efficient market return is the best linear minimum variance estimator (BLMVE) of the security’s mean return (in the sense that it minimizes the sum of the traders’ mean squared forecast errors). Thus, an efficient market aggregates traders’ heterogeneous information in an optimal way. Fourth, we prove that an efficient market produces a mean return (price) as a Blackwell sufficient (most informative) experiment among all possible aggregated expected return (price) forecasts.