[Machines produce a constant flow of output when operating, and break down in a random process. The task of workers is to repair the broken machines. Servicing time is also a random process [2]. The limiting steady state probabilities and the relation between output and inputs are derived, and fitted to a Cobb-Douglas production function.]
In this paper a distribution sampling study consisting of four major experiments is described. The L1 norm is employed in two new estimating techniques, direct least absolute (DLA) and two-stage least absolute (TSLA), and these two are compared to direct least squares (DLS) and two-stage least squares (TSLS). Four experiments testing the normal distribution case, a multicollinearity problem, a heteroskedastic variance problem, and a misspecified model were conducted. Two small sample sizes were used in each experiment, one with N = 20 and one with N = 10. In addition, conditional predictions were made using the reduced form of the four estimators plus two direct methods, least squares no restrictions (LSNR) and another new method known as least absolute no restrictions (LANR). The general conclusion was that the L1 norm estimators should prove equal to or superior to the L2 norm estimators for models using a structure similar to the overidentified one specified for this study, with randomly distributed error terms and very small sample sizes. BEGINNING WITH the method developed by Haavelmo [11] for solving the problem of single equation bias, econometricians have devoted considerable effort to developing additional methods for estimating the structural parameters of simultaneous equation models [2,12,20,24]. While it has been fairly easy to develop the asymptotic properties of these estimators, a distinguishing characteristic of econometric models is that they are invariably based upon small samples of data and thus, the asymptotic properties of the various estimators are not
[In recent years spectral techniques have been used to assess the effects of applying various types of seasonal adjustment procedures to economic time series. Similar analyses using artificially generated time series have also been attempted. The effects, desirable or undesirable, of a particular method of seasonal adjustment can, however, only be assessed properly in the time domain and only in relation to the objectives of such adjustment. Despite the fact that such objectives have not been clearly formulated nor any definitive conception of the nature of seasonality developed, in this paper we do adopt a general approach consistent with what has been written on the subject since the time of Jevons. In terms of a simple three component model of an economic time series having properties similar to those found in many actual time series, we devise several "methods" of seasonal adjustments based on a minimum mean-square-error criterion of optimality. We show that such methods of seasonal adjustment produce seasonally adjusted series bearing the same relationship to the unadjusted series in spectral terms as that found by Nerlove and others in their studies of BLS and Census methods of adjustment. Our conclusion is not that spectral methods are useless, but rather that comparisons in the frequency domain must be interpreted with great care. Further research must emphasize objectives and models. Whether these are formulated in frequency terms or in the time domain is of secondary importance.]
Drawing up a medium term economic plan usually involves a complicated interaction between the planning ministry and representatives of the various industries, firms, or departments. Each economic agent works in his own environment with at best incomplete information about the other agents. Yet somehow the economic system as a whole is typically able to move toward an operational plan which is satisfactory even when judged by the criterion of complete information. This paper examines the properties of one particular theoretical model of economic planning in which the center transmits information via a system of quotas.