Fama, Eugene, and Richard Roll, Some Properties of Symmetric Stable Distributions, Journal of the American Statistical Association 63 (Sept. 1968), 817-838. ______ Parameter Estimates for Symmetric Stable Distributions, Journal of the American Statistical Association 66 (June 1971), 331-338. Friedman, Daniel, and Stoddard Vandersteel, Short-run Fluctuations in Foreign Exchange Rates: Evidence from the Data 1973-79, Journal of International Economics 13 (Aug. 1982), 171-186. Gnedenko, B. V., and A. N. Kolmogorov, Limit Distributions for Sums of Independent Random Variables (Reading, MA: Addison-Wesley, 1968). International Monetary Fund, International Monetary Fund Survev, various issues. Judge, George G., William E. Griffiths, R. Carter Hill, and Tsoung-Chao Lee, Theory and Practice of Econometrics (New York: John Wiley and Sons, 1980). Kendall, Maurice G., and Alan Stuart, Advanced Theory of Statistics, 2nd Edition, Vol. I (London: Charles Griffin and Co., 1963). Kindleberger, C. P., The Benefits of International Money, Journal of International Economics 2 (Sept. 1972), 425-442. Mandelbrot, Benoit, Variation of Certain Speculative Prices, Journal of Business 36 (Oct. 1963), 394-419. McFarland, James W., R. Richardson Pettit, and Sam K. Sung, The Distribution of Foreign Exchange Price Changes: Trading Day Effects and Risk Measurement, Journal of Finance 37 (June 1982), 693-715. Moulton (Westerfield), Janice, An Estimation of Foreign Exchange Risk Under Fixed and Floating Rate Regimes, Journal of International Economics 7 (June 1977), 181-200. Rogalski, Richard J., and Joseph D. Vinso, Empirical Properties of Foreign Exchange Rates, Journal of International Business Studies 9 (Fall 1978), 69-79. Roll, Richard, Behavior of Interest Rates (New York: Basic Books, Inc., 1970). Schmidt, Peter, Econometrics (New York: Marcell Dekker, Inc., 1976).

This paper presents estimates of the rate of technical change in the electric power industry over the period 1951-78. The estimated model directly incorporates the effects of rate-of-return regulation, and uses both a time trend and a vintage index to represent disembodied and embodied technical change, respectively. The results indicate that disembodied technical change was the primary source of cost reduction during 1951-70, and that tighter regulation, as represented by a one point reduction in the rate of return, would have reduced the rate of technical change by an average of 1%-2% during 1951-78. 34 references, 18 footnotes, 3 tables.

The preserving of flexibility when faced with uncertainty is a neglected aspect of behaviour under risk. Yet it is an important factor in decisions to hold liquid assets or delay irreversible investment. This paper formalizes the notion of flexibility in a sequential decision context, and relates its value to the amount of information an agent expects to receive. A rudimentary money demand model is developed embodying these ideas, and the history of flexibility as an economic concept is traced.

M. Laurentius Marais, James M. Patell, Mark A. Wolfson, The Experimental Design of Classification Models: An Application of Recursive Partitioning and Bootstrapping to Commercial Bank Loan Classifications, Journal of Accounting Research, Vol. 22, Studies on Current Econometric Issues in Accounting Research (1984), pp. 87-114

* Professor, Cornell University; t Professor, Rice University. We wish to thank our colleagues and particularly Professors William H. Beaver and Joel Demski, both of Stanford University, and Robert J. Swieringa, Cornell University, for their careful reading of this long manuscript and the substantive improvements they suggested. The limitations and omissions remain our responsibility. 1 Given at the 1970 Annual Meeting of the American Accounting Association, the poem, if such is a valid description, was composed by the first editor of JAR, David Green, and published in JAR [1970] with appropriate disclaimers.

In this paper we provide evidence on the relationship between revisions in analysts' composite earnings forecasts and contemporaneous stock price movements. Most of the research on forecasts of accounting earnings has focused on (1) whether earnings forecasts have information content, and (2) the pros and cons of regulations (standards) requiring earnings forecasts (Gonedes, Dopuch, and Penman [1976]). The second issue is moot if earnings forecasts do not contain information. Empirical evidence suggests that management earnings forecasts are associated with significant security price revisions (Foster [1973], Patell [1976], Nichols and Tsay [1979], and Penman [1980]). Similar results were obtained for analysts' forecasts (Gonedes, Dopuch, and Penman [1976]). One aspect of this area of research which has been neglected is the relationship between analysts' forecasts and managements' forecasts. Normally, corporations do not issue public forecasts of accounting data on a regular basis.' However, security analysts are in frequent contact with corporations in an effort to confirm information or obtain new

This paper examines the effects of Broad Tape news releases of earnings and dividend announcements on three aspects of intraday stock price behavior: mean returns, return variance, and serial correlation in consecutive price changes. The initial price reaction is evident in the first pair of price changes following the release (i.e., within a few minutes, at most). The returns earned by simple trading rules dissipate within five to ten minutes, although significant returns are detected in the overnight period and at the opening of trading on the next day. Disturbances in the variance and serial correlation persist for several hours and extend into the following trading day. As a class, dividend announcements induce much less activity than do earnings, although the response to dividend changes is comparable to the earnings announcement effect.

Pseudo maximum likelihood techniques are applied to basic Poisson models and to Poisson models with specification errors. In the latter case it is shown that consistent and asymptotically normal estimators can be obtained without specifying the p.d.f. of the disturbances. These estimators are compared both from the finite sample and the asymptotic point of view. Quasi generalized PML estimators, which asymptotically dominate all PML estimators, are also proposed. Finally, bivariate and panel data Poisson models are discussed. THE ANALYSIS OF ECONOMIC BEHAVIOR often leads to the study of characteristics taking a small number of positive values. The classical linear model is not adapted to explain how such discrete variables depend on other quantitative or qualitative variables. The reasons are similar to those usually given in the case of an endogenous qualitative variable: the shape of the observation set does not correspond to a linear model, the assumption of normality of the disturbances cannot be made, since the endogenous variables take a small number of values with strictly positive probabilities, and the prediction formulae which are deduced from a linear model give impossible values. In the models considered in the literature to describe discrete variables (Cox and Lewis [2], El. Sayyad [3], Frome, Kutner, and Beauchamp [4], Gilbert [5], Hausman, Hall, and Griliches [8]; see also Lancaster [12]) the endogenous variable is assumed to have a Poisson distribution conditional upon the exogenous variables. The parameter of this distribution is a function of the values of the exogenous variables. The choice of such a model is justified if the dependent variable counts the occurrence of a given event during a fixed period and if the usual assumptions of the Poisson process are satisfied. For instance, the model is adapted to describe daily numbers of oil tankers' arrivals in a port, the number of accidents at work by factory, or the number of patents applied for and received by firms (Hausman, Hall, and Griliches [8]).