This paper is devoted to the formulation of tentative formal models on dependency in both political and economic dimensions, and to estimate them with empirical data on Latin America, the home ground of the dependency theories. In the political dimension, the model tries to “explain” the coercive authoritarianism of the regime by employing the concept of political dependence. In the economic dimension, developmental performance and economic inequality are explained by the components of economic dependence, such as capital and trade dependence. Explanatory constructs of the models include not only those “dependency” variables, but also the internal domestic factors. The explanatory power of the models are surprisingly high, and most of the key coefficients are statistically significant. Residuals are also examined.
A fundamental idea of a new algorithm of minimum dimension analysis of ordered class belonging (MDA-OR) which is one of the methods for nonmetric multidimensional scaling (MDS) is described in the author's previous paper (1978). But the previous paper is mainly concerned with one dimensional case. In the present paper, the author gives a full account of a new algorithm together with its computational techniques, and shows some numerical examples in multidimensional case.
Experiments were carried out on the pitch of two groups of computer-generated complex tones. The sequence of the ten complex tones in group A are perceived as an endless scale when they are heard repeatedly (Shepard, 1964). Complex tones in group B were synthsized by slight modification of frequency structure of complex tones in group A. The experimental results were analyzed by eij-type quantification theory. The configuration of ten complex tones in group A is considerably circular. On the other hand, the configuration of ten complex tones in group B shows a considerably onedimensional character. This difference in configuration was explained by the place cue and the time cue originated from neural activities of auditory neurons.
We describe a convergent procedure for fitting the common factor analysis model to multivariate data whose variables may be nominal, ordinal or interval. Any mixture of measurement levels is permitted. There may be any pattern of missing data. As distinguished from previous work, the nonmetric relations (nominal or ordinal) are assumed on the raw observations (not on the correlations), and the model fitted is the common factor analysis model (not the principal components model) which isolates common from unique factor variation. The computational algorithm, based on the alternating least squares principle, is monotonically convergent and efficient. An illustrative example is presented.
“The situational decision making model” is a qualitative, non-metric approach to a decision making. Originally aimed at a more realistic application of the statistical decision theory, the model does not assume an assumptive loss function, but consists of more essential ingredients of a decision making, i.e., ‘decision criteria’, ‘situations’ and ‘actions’. The paper is of a cognitive nature, dealing with how to order, retrospectively, the decision criteria in terms of their influences, and hence to analyse the structure of decisions.