"... In PAGE 8: ... A transition matrix can be constructed from the concept history and dynamically updated upon each future occurrence of concept change. Table3 shows the current status of a flrst-order transition matrix4. Suppose that the current concept is autumn.... In PAGE 12: ...equivalence measure as in Table 1, a concept history C HIS learned as in Table 2, a transition matrix C TRA learned from C HIS as in Table3 , a window of trig- ger instances I WIN, a probability threshold thresholdprob, a classiflcation accuracy threshold thresholdaccu Output: a concept (prediction model) cnext for oncoming instances Begin clast = the last stable concept in C HIS; // Proactive cnext(s) = concept(s) whose probability given clast is bigger than thresholdprob, ac- cording to C TRA; IF multiple cnexts exist // Break tie by historical-reactive FOREACH cnext calculate its accuracy on I WIN; cnext = the one acquiring the highest accuracy; // Historical-reactive IF no cnext exists FOREACH concept chistorical 2 C DIS calculate its accuracy on I WIN; cnext = the one acquiring highest accuracy; // Contemporary-reactive IF accuracy of cnext on I WIN is less than thresholdaccu cnext = concept learned from I WIN; return (cnext); End Table 4. RePro: reactive+proactive attribute values and measures the conceptual equivalence by syntactical com- parisons.... ..."

"... In PAGE 11: ...25%) have more than 30 spells, two have more than 40 spells, and one household has 87 spells. Table1 tabulates the frequency counts of the l to m transitions for each of the 6 6 transitions... In PAGE 12: ... The mean interpurchase time for the whole sample is about 55 days. (insert Table1 about here) (insert Table 2 about here) One of the objectives of our study is to investigate the relationship between the pattern of brand switching and interpurchase times. A comparison of the empirical distributions for the interpurchase times between transitions should provide some insights.... ..."

"... In PAGE 7: ....e. the time taken to complete 11 ANs The hypothesis, concerning convergence of tracks, is tested by comparing the transition matrixes of both groups. Table2 and 3 show the transitions of both the control group and the experimental group (the course modules numbered from 1 to 11). ... ..."

"... In PAGE 5: ... Nevertheless, the markovian framework is a useful one for summarizing the data provided the statistics presented are interpreted as describing the outcomes of the forces governing the intergenerational evolution of attainment during the 1960s, 70s, and 80s. Table1 presents the estimated transition matrix and the implied equilibrium distribution. The probability of transition from category to category is estimated by the quasi-maximum likelihood estimator where is the sum of the weights G01G02 G01G02 G01G02 G01G02 G02G03G04 G05 of those father-son pairs who transition from category to category for The G09 G0A transition matrices estimated here satisfy the conditions under which the equilibrium distribution exists given in Feller [1970].... ..."

"... In PAGE 5: ... The probability of transition from category to category is estimated by the quasi-maximum likelihood estimator where is the sum of the weights G01G02 G01G02 G01G02 G01G02 G02G03G04 G05 of those father-son pairs who transition from category to category for The G09 G0A transition matrices estimated here satisfy the conditions under which the equilibrium distribution exists given in Feller [1970]. The equilibrium probability of being in category is for , found as the solutions to for , and G09 G0A G09 G0D G01G01G02G01 G02G03G04 G05 G01G03G04 G05 G01 Table2 shows the estimated transition matrix and equilibrium distribution for the income data. This data is adjusted for measurement errors, cohort effects, the different positions of the fathers and sons in the life cycle, and racial differences in the age-earnings profile.... ..."