APPLICATIONS OF DECISION TREES TO THE EVALUATION OF APPLIED RESEARCH PROJECTS

Abstract I n this paper two applications of decision trees are described. These relate to the electronics and chemical industries and are concerned with the evaluation of applied research projects aimed at new products. It is demonstrated that decision tree diagramming is a demanding yet flexible technique which allows the representation of sequential decisions and subjectively based data in a readily understood form.     

Estimating the covariance matrix for regression models with ar(1) errors and lagged dependent variables

We propose a simple procedure, based on an artificial linear regression, for consistently estimating the covariance matrix of the parameter estimates for linear regression models with serially correlated errors and lagged dependent variables.     

VOTING AND THE INTERTEMPORAL SELECTION OF TAX RATES IN A MACRO-ECONOMY*

This paper presents a macroeconomic model where governments are motivated both by their ideological preference for alternative fiscal policies that vary in their short run and long run consequences and by their reelection prospects. I prove that a sequential equilibrium exists in which moderate incumbents compromise their ideologies in order to get reelected, while extreme governments forego reelection but tie the hands of their successors. The view implied by this model is that incumbent governments, in order to get reelected, do not ideologically differentiate themselves from one another since such behavior is likely to result in loss of office. Alternatively, governments that are either subsequently not reelected or are lame ducks pursue ideologically preferred policies since they are less constrained by the electorate.     

Foreign owners and American cement: Old cartel hands,or new kids on the block?

In 1961, all portland cement producing capacity in the United States was owned by companies domiciled here. Thirty years later, about 70% was controlled by foreign owners, mostly European. It has been suggested by various authorities — academic economists, the business press, even cement executives — that foreign governments' tolerance for formal or tacit collusion would bring the American cement industry a kind of ‘price discipline’ that it has been largely unable to practice since 1948. This contention is tested as an hypothesis, using regional data from the U.S. Bureau of Mines and the Portland Cement Association. The test shows either no or a negative relation between foreign control and the level of regional cement prices. At the same time, we observe the traditional positive relation between concentration (HHI) and prices. The results imply that horizontal mergers between cement producers are a better subject for the attention of public policy than where they are owned.     

OVER-REACTIONS IN US AGRICULTURAL COMMODITY PRICES

The purpose of this paper is to test rationality in agricultural commodity markets. We investigate the price-adjustment process after significant events using abnormally large cash commodity price changes as proxies for the arrivals of significant events in the markets. The evidence suggests that market rationality is violated. Generally, agricultural commodity prices tend to reverse after significant events. This is consistent with the over-reaction hypothesis which maintains that traders in spot commodity markets over-weight more recent information and under-weight prior information in their expectations.     

Capitalist lessons for communists

Are there any tips that capitalists can offer former communists in introducing and managing a market economy? Russell Lewis reviews the dangers and opportunities.     

Forecasting Housing Approvals in Australia: Do Forecasters Herd?

Price trends in housing markets may reflect herding of market participants. A natural question is whether such herding, to the extent that it occurred, reflects herding in forecasts of professional forecasters. Using more than 6,000 forecasts of housing approvals for Australia, we did not find evidence of forecaster herding. On the contrary, forecasters anti‐herd and, thereby, tend to intentionally scatter their forecasts around the consensus forecast. The extent of anti‐herding seems to vary over time. We also found that more pronounced anti‐herding leads to less accurate forecasts.