Affiliation:
1. Netherlands Economic Institute, Rotterdam
Abstract
According to the multiple-gap hypothesis, investment decisions, determined by the joint action of multiple determinants, result from the dominance at the time of the decision of the most constraining of these factors. This hypothesis is tested on Dutch investment data for the period 1951-1974 by means of three types of models: a "rigid" model in which a single regime is selected at any time, and two "flexible" models allowing for a joint action of the various determinants, the last model incorporating a switching mechanism with respect to the dominant factor. Estimation problems are discussed and the results obtained with the different models are compared.
Reference22 articles.
1. BROWN R.L. et DURBIN (1968), "Methods of investigating whether a regression relationship is constant over time", Paper presented at the European Statistical Meeting, Amsterdam.
2. CHOW, G. (1960), "Tests of the quality between two sets of coefficients in two linear regressions ", Econometrica, 28, pp. 561-605.
3. COX, D.R. (1962), "Further results on tests of separate families of hypotheses", Journal of the Royal Statistical Society, 24, Series B.
4. DEN BROEDER, G., "Between Newton and Cauchy: the diagonal variable metric method; theory and test results", to appear in Journal of Applied and Computational Mathematics.
5. EISNER, R. (1977), "Cross section and time-series estimates of investment functions", Annales de l’INSEE, 30-31, pp. 99-130.