This study demonstrates that the mixed diffusion-jump process is superior to the stable laws or a mixture of normals as a model of exchange rate changes for the British pound, French franc, and the We st German mark relative to the United States dollar. The parameter value s for the mixed diffusion-jump process are dependent on the monetary policy regime in force in the United States, with the estimates for the franc and mark being intertemporally similar but different from the pound. Copyright 1988 by MIT Press.

[The relation between theorized components of the bid-ask spread and trade size for a sample of NYSE firms is examined. We find that the adverse selection component increases uniformly with trade size. Conversely, order processing costs decrease with increases in trade size for all but the largest trades. We find that order persistence decreases with trade size. The adverse selection component is highest at the beginning of the day and lowest at the end of the day for all but the largest trades. Trades of NYSE firms executed on regional exchanges or NASDAQ contain a large order processing cost component but no significant adverse information effect.]

We provide empirical evidence on the economic benefits of negotiating trades in the upstairs trading room of brokerage firms relative to the downstairs market. Using Helsinki Stock Exchange data, we find that upstairs trades tend to have lower information content and lower price impacts than downstairs trades. This is consistent with the hypotheses that the upstairs market is better at pricing uninformed liquidity trades and that upstairs brokers can give better prices to their customers if they know the unexpressed demands of other customers. We find that these economic benefits depend on price discovery occurring in the downstairs market.

Recently a number of mathematical programming models have been developed to assist banks in their portfolio (balance sheet) management decision making. Generally, the model structures used may be classified as either linear, linear goal, or two-stage linear programming. Of these, linear programming models are the most common. The purpose of this paper is to discuss the optimal bank portfolio management solutions produced by each of the above programming structures. In addition, a new model structure, two-stage linear goal programming, is developed and compared to the other structures. From a decision-making perspective, this new model structure is found to provide additional useful information.