### Table 5. Uop Linear Regression Model Comparison

2005

"... In PAGE 3: ... Therefore, a more accurate power model should use the number of uops fetched per cycle instead of the number retired. Table5 provides a comparison of linear regression power models based on the three previously mentioned metrics. Average Uops / Cycle 0.... ..."

Cited by 9

### Table 5: FFED Linear Regression Model

in ABSTRACT A CASE STUDY IN META-SIMULATION DESIGN AND PERFORMANCE ANALYSIS FOR LARGE-SCALE NETWORKS

"... In PAGE 7: ...306 1, as in the RRS experiments. In Table5 , we fitted a regression model to the data and found that the same input parameters had the most effect on the model. FFED model produced an adjusted R-squared value of We were confident that RRS was properly modeling parameter space as we would have explored had a more detailed FFED.... ..."

### Table 5: Parameters of linear regression models

"... In PAGE 8: ... Financial data for 2001 has not been taken into account, because correlation analysis has not indicated any information about the expected relationships between variables. Table5 . shows the results of tested regression models based on their crucial characteristics: coefficient of determination R2 (indicating how much of the variance between predicted and actual values does the model actually explain); adjusted coefficient of determination R2 (deemed as a more conservative estimate than the previously described one); F-test significance (which indicates the statistical significance/acceptability of the entire model); and colinearity index (whose value may indicate the problem of strong correlation of independent variables, which would be a significant violations of the assumptions behind the statistical model).... ..."

### Table 2: Results for ANN and linear regression models

### Table 1. Multiple linear regression models

"... In PAGE 2: ... DIAGNOSIS: The predictive value of the clinical data in OSAS diagnosis is low. Hoffstein [5] published results that indicated that clinical data explains 36% of the variability of the AHI (apnea hypopnea) and Katz [6] reported a figure of 39%, other authors report lower figures ( Table1 ). The subjective clinical evaluation of the interviewer has also been evaluated and tends to have a low sensibility and specificity, in the order of 55%-65% respectively, for correctly classifying the sick.... In PAGE 4: ... The linguistic labels are considered as being in ascending order, from left to right. Thus we can meaningfully assign ordered ascending integer values thus: {0,1,2,3,4} We have chosen a subset of variables from the questionnaire which in the literature (see Table1 ) have been identified as the most discriminatory variables with respect to apnea diagnosis. These are: age, sex, weight, body mass index, neck circumference, alcohol intake, blood pressure, snoring and daytime sleepiness.... In PAGE 5: ...ariables (age, neck circumference, etc. ..) and 8 question responses (from a total of 41). These variables are listed in Table 2. We have used expert medical knowledge, results from the literature ( Table1 ) and statistical analysis techniques to select the key variables. In aggregating we have considered all data as numeric.... ..."

### Table 2 Results for ANN and linear regression models Models ANN model Linear Regression model

### Table 6: Linear Regression Models Sequence The Best Fit Model

"... In PAGE 5: ... Linear regression is used to fit linear models to the estimates of the expected error values as function of the observation number for van der Corput sequences in base 2,3,4, and 5 respectively. Table6 summarizes the outcomes, and shows that placing observations as an S5 sequence will be as efficient as 98% compared to S2. Table 6: Linear Regression Models Sequence The Best Fit Model... ..."

### Table 5. Forecast accuracy of the multiple linear regression model

"... In PAGE 7: ... If the algorithm is not able to detect any patterns in the training set a recognition rate of 33% on average can be expected. Table5 illustrated the recognition performance of the multiple linear regression model. ... ..."

### Table 4: Parsimonious linear regression models for High School data

"... In PAGE 13: ... 14 This step was particularly important, given that no predictors were revealed as statistically significant in the full linear regression models. Table4 displays three alternative parsimonious models for the high school data set. Each model contains a proxy for socio-economic level of the school environment, either parent level of education or drop-out rate.... ..."

### Table 2: RRS Linear Regression Model Output generated by R.

2004

"... In PAGE 6: ...07 seconds. We fitted a linear regression model to our data using a tool called R [25], and generated the co-efficients shown in Table2 . After analyzing the variance on the inputs, we found the parameters that had the greatest impact on the model to be the HELLO packet interval, HELLO inactivity timer, and the SPF computation interval.... ..."