### Table 2: Development of stem overregularization errors

"... In PAGE 8: ... To investigate developmental changes, the various recordings were assigned to stages of development defined in terms of the mean length of utterance (MLU); see Brown (1973). Table2 shows a breakdown of the stem overregularization errors against the number of correct marked stem forms across these stages. Table 2: Development of stem overregularization errors... In PAGE 8: ... Table2 shows that the period from stage II onwards in which stem overregularization errors frequently occur is preceded by a period (= Stage I) in which almost all of the required marked stem forms are correctly produced by the children. Moreover we can see from Table 2 that in the age period represented in the longitudinal data, there are no signs of stem overregularization errors disappearing from the speech of the children or decreasing over time.... In PAGE 8: ...Table 2 shows that the period from stage II onwards in which stem overregularization errors frequently occur is preceded by a period (= Stage I) in which almost all of the required marked stem forms are correctly produced by the children. Moreover we can see from Table2 that in the age period represented in the longitudinal data, there are no signs of stem overregularization errors disappearing from the speech of the children or decreasing over time. Consider now the experimental results from the older children.... ..."

### Table 4--Infrequency of Purchase Model

"... In PAGE 19: ... T he superiority of the log-infrequency of purchase of model relative to the other three models is mani fested in other non-statistical dimensions. For example, the frequency of purchase mode l estimates in Table4 imply a sample average probability of negative consumption of postal services of 18.4%, which is clearly an unreasonable implication of this model.... ..."

### Table 1 shows that corruption is quite infrequent,

1996

Cited by 97

### Table 5--Log Infrequency of Purchase Model

"... In PAGE 22: ... Th e remaining parameters in the consumption share equation and the purchase probability equation are a ssumed to be fixed over time. For these variables, the resulting parameter estimates are very similar to those repor ted in Table5 . There are changes in the coefficients on log-prices, log-total non - d urable expenditure and probability of computer ownership variables that reflect the fact that we have i ncluded these variables interacted with a time trend in the model.... ..."

### Table 6--Infrequency of Purchase Model with Corner Solutions

### Table 11--Log Infrequency Model with Time Trends

### Table 3 shows a clear frequency effect in the stem overregularizations for both classes of strong verbs: low-frequency stems elicit significantly more stem errors than high- frequency ones (t(25) = 10.399, p lt;.001). Finally, in order to examine developmental changes, we plotted in Figure 1 the percentages of overregularizations against age. A regression analysis revealed that stem overregularizations are significantly linked to age and that the overregularization errors gradually decrease with age (R2=.174, f(x)=73.51-5.243x, F(1,24)=6.252, p lt;.05). Figure 1: Stem overregularizations in relation to age

"... In PAGE 8: ...n cases in which strong stem forms are required in the adult language, i.e. errors such as those in (5) above. Table3... In PAGE 9: ... Table3 : Elicited stem forms Required stem form and stem frequency Number of stem errors Number of correct stems Error percentages -i- / high frequency 19 136 12.2% -i- / low frequency 89 59 37.... ..."

### Table 3. Classi cation results of all infrequent diagnoses together.

"... In PAGE 5: ... The combined average PSM avg(Bayes, Heuristic, Case{based) compensates best the weakness of the individual PSM. The classi cation result is of all in- frequent diagnoses ( Table3 ) the highest. This method combines the highest recognition rate (45.... In PAGE 6: ... Classi cation results of all infrequent diagnoses together. Bayes) acts for the infrequent diagnosis ( Table3 ) and the Non{speci c abdomi- nal pain cases (Table 5) like a Naive Bayes classi cator and for the Appendicitis cases (Table 4) like a Heuristic classi cator. This is a direct consequence of the fact, that the diagnostic knowledge was provided exclusively by an expert.... ..."

### Table 4: Reducing False Positives II: use of infrequent substrings in ngerprints.

1996

"... In PAGE 7: ... They are, however, su - ciently di erent to be useful in tandem. Table4 considers the e ectiveness of focusing on... ..."

Cited by 1

### Table 5--Purchase Probability Derivatives for Log Infrequency Model

"... In PAGE 16: ...32, many households have estimated purchase frequencies above and below this value. Table5 gives the probability derivatives associated with the postage purchase probability for this model. This table shows that increases in the probability of personal computer ownership significantly reduces the purchase frequency of postal delivery services.... ..."