Knowledge that Transforms

Maxwell L. King, Thomas S. Shively, Locally Optimal Testing When a Nuisance Parameter is Present Only Under the Alternative, The Review of Economics and Statistics, Vol. 75, No. 1 (Feb., 1993), pp. 1-7

The demand for two custom calling services is investigated. The services may be bought individually or in a discounted package. A micro-theory based on discrete choice model is formulated that explicitly accounts for these purchase options. The model is estimated assuming both dependence and independence of the unobservable choice-influencing variables. The estimated parameters are used to simulate the revenue impact of price and discount changes.

In this paper, the time-varying parameter model based on the Kalman filter is combined with the binary choice model. Next, estimation of the unknown parameters is examined without using any distribution. Finally, a money excess demand function is estimated as an application to the problem.

David N. DeJong, Steven Husted, Towards a Reconciliation of the Empirical Evidence on the Monetary Approach to Exchange Rate Determination, The Review of Economics and Statistics, Vol. 75, No. 1 (Feb., 1993), pp. 123-128

The authors reply to the comment by D. R. Kamerschen and J. Park on their 1988 paper published in this Review. They find that the econometric point raised by these authors is flawed, because differences in model structure and data are ignored. In particular, the importance of materials input in assessing price-cost margins is reiterated here and illustrated with the 1988 paper's original table. Other points of the comment are refuted by direct reference to statistical results and inferred conclusions in the 1988 paper.

R. Ian Domowitz, Glenn Hubbard, and Bruce C. Peterson (1988) estimate industry markups of price over marginal cost and show the influence of market structure on cyclical movements in total factor productivity, using a disaggregate panel data set of 284 four-digit, Standard Industrial Classification industries. In this comment, the authors examine whether Domowitz, Hubbard, and Peterson really estimate what they intend to estimate. Domowitz, Hubbard, and Peterson also claim their 'estimates are not as large as those of Hall.' If it is not clear what they estimate, the comparison is of limited usefulness. Nevertheless, the Domowitz, Hubbard, and Peterson study deserves consideration and contemplation by anyone interested in industrial economics.

This Monte Carlo study compares the performance of a recently proposed multiprocess mixture model and a more traditional random walk time-varying parameter model in the face of structural shifts and outliers. The mixture model performs well and the latter model performs poorly. This finding is of general interest since investigators often adopt random-walk time-varying parameter models to accommodate potential regime shifts in regression relationships. The findings suggest that the time-varying parameter estimation procedure is unlikely to find abrupt shifts, since the time-varying parameter estimates are contaminated by the outliers and regime shifts.

This paper shows that U.S. monthly consumption data are consistent with the permanent income hypothesis when transitory consumption and temporal aggregation effects are jointly incorporated into the model. In this case, a more appropriate representation for the permanent income hypothesis is the integrated-moving average IMA(1, 1) process with a negative MA coefficient, rather than the repeatedly rejected random walk process. Restrictions on the relative importance of transitory and permanent consumption are also discussed, with and without measurement errors.

This paper extends the work of Boyd (1987) by examining the question of efficiency in Yugoslavian agricultural production using the stochastic production frontier. We find the private sector produces with higher output efficiency than the social sector. Next, we examine regional efficiency differences. Our findings reinforce earlier analysis of the economic impact of regional development policy pursued in Yugoslavia. Less developed republics appeared unable to utilize efficiently the large volume of investment resources allocated from the more developed republics via the central government.

This paper provides improved estimates of technological change and the first estimates of factor bias in Polish industry. The analysis uses newly generated data on Western capital (the factor assumed responsible for technical change), adjusts capital for capacity utilization, and tests for more complex specifications of the production function. The major findings are (1) the trend in technical change was nonincreasing over the period and technological change was Western-capital-using in only two industries (suggesting the government misallocated resources), and (2) technical change was nonneutral in six of the eight industries (suggesting the existing literature suffers from misspecification).