The relationship between an investor's attitude toward risk and the shape of his preference functions has long been recognized in both the general portfolio problem and the mean-variance model. By contrast, the literature has largely ignored the connection between general measures of an investor's attitude toward risk and the shape of his mean-variance or mean-standard deviation indifference curves. Yet this relationship is significant. Through general measures of risk aversion, assumptions about an investor's behavior under uncertainty imply restrictions on indifference curves. Conversely, assumptions about indifference curves impose restrictions on an investor's behavior under uncertainty. The development of this relationship and its implications is the objective of this note.

Assuming continuous trading in continuous time with Brownian motion processes, the basic capital asset pricing model of Sharpe, Lintner, and Mossin is developed under arbitrary distributions of investors' beliefs consistent with available information. Results on the processing of information are reported, and properties of investors' portfolios are derived.

The distributions of the LIML and TSLS estimates of the coefficient of an endogenous variable in a single equation can be approximated by asymptotic expansions. This paper relates the expansions in terms of the noncentrality parameter and the sample size going to infinity, the noncentrality parameter going to infinity with the sample size held fixed, and the standard deviation of the disturbance going to zero (small-o). 1. INTRODUCriON RECENTLY, ASYMPTOTIC EXPANSIONS of the distributions of estimates of coefficients of a single equation in a system of simultaneous equations have been made by Anderson [1], Anderson and Sawa [2], Mariano [6 and 7], and Sargan and Mikhail [11]. The expansions have usually been carried out on the basis that the sample size increases and that the effect of the exogenous variables (the noncentrality parameter) increases along with the sample size. In this paper we consider the case of the covariance matrix of the disturbances known and alternatively the case of the sample size fixed. We relate these three cases to the approach of letting the disturbance decrease (the small-o- approach). The estimates treated are two-stage least squares (TSLS) and limited information maximum likelihood (LIML).

T HERE has been much theoretical work done on models of information and search beginning with the work of Stigler (1961, 1962) but there has been little empirical investigation of the implications of these models. With the importance these models have attained in describing macroeconomic phenomena such as the Phillips curve, this empirical work is necessary to guide any potential methods designed to reduce the unemployment rate. This paper examines the time path of wage demands of the unemployed as a test of some of the implications of the search models. Most of the theoretical work has been devoted to analyzing the optimal behavior of individuals who must make choices on the basis of incomplete information and of the equilibrium behavior of markets whose participants behave according to particular search rules. For labor markets it follows that it is not necessarily optimal for an individual to accept the first job offered to him, and thus, the equilibrium position will be characterized by positive unemployment. The search strategy of an unemployed individual usually takes the following form. Search until a wage offer is received that is above some reservation wage, this reservation wage being determined by maximizing expected returns. Since search is a sequential process, the sequence of reservation wages completely describes the behavior of the agents.1 Furthermore, it is derived in most of the theoretical work that this sequence of reservation wages is either constant or monotonically declining. People who remain in the market are willing to accept successively lower wages as time passes.2 This seems to be a paradoxical result about learning, i.e., time always makes one pessimistic. The models in which this monotonicity property is generally derived, do not consider learning as part of the mechanism generating behavior, but for any consistent model of both search and turnover (implicit in the search theories of the Phillips curve) it is required that individuals revise upward their expectations of the wage distribution with the state of the economy. In an economy that is constantly changing, learning should be an important determinant of search. It is hypothesized that when one is permitted or required to learn about the wage distribution through sampling, it seems reasonable to expect that initially pessimistic individuals will revise their wage demands upward before sampling terminates. This paper will argue that the above hypothesis is correct and that the sequence of reservation wages is not monotonically declining. The first part of the paper will present a heuristic formulation of a search and learning model where it can be seen that the sequence of reservation wages depends on the initial expectations of an individual and on the particular sequence of information (including wage offers) that an individual obtains. Although the particular search rule analyzed is not derived from optimization, it should help develop the intuition necessary for believing that reservation wages can and do rise in the course of search. The formulation could be considered as the study of behavior characterized by bounded rationality, but it is mainly presented to motivate the empirical work. The empirical evidence presented supports the hypothesis that a monotonically declining sequence of reservation wages is not an accurate description of actual search behavior of unemployed individuals looking for jobs. The sequence of reservation wages depends heavily on the perceived and actual wage distribution.

Arthur T. Denzau, Amoz Kats; Expected Plurality Voting Equilibrium and Social Choice Functions, The Review of Economic Studies, Volume 44, Issue 2, 1 June

The efficient market, martingale model of security price movements requires that the arrival of new information be promptly arbitraged away. A necessary and sufficient condition for the existence of an arbitraged price is that statistical dependence among prices must decrease very rapidly. If persistent statistical dependence is present, the arbitraged price changes do not follow a martingale and should have an infinite variance. Using a technique for detecting long-term dependence, called R/S analysis, 200 daily stock return series are studied; many series are characterized by long-term dependence. Thus, in the presence of long-term dependence, the martingale model does not hold. Also, the distribution of security returns is non-normal stable Paretian as opposed to Gaussian.