### Table 1. Summary of scalar shape-of-motion descriptors. Summations are over the entire image. u and v are the x- and y-direction op- tical flow values respectively. The function T segments the image. T = 1 for pixels that are moving and T = 0 for stationary pixels. s are eigenvalues of second moment matrix for motion distribution.

1997

"... In PAGE 3: ... Some of these are locations in the image, but all are treated as time-varying scalar values. Table1 summarizes the scalar values used. Rearranging the scalar values forms a time series for each scalar.... ..."

Cited by 8

### Table B.2: The simple perceptron mean eld algorithm with self-averaging second moments of the cavity eld, = . B is the correlation matrix of the input distribution p(s).

1998

Cited by 6

### Table 1. Condition numbers of the matrix of periodized second derivative (with and without preconditioning) in the system of coordinates associated with Daubechies apos; wavelets with three vanishing moments M = 3.

### Table 2: Second moment of service time

2005

"... In PAGE 11: ...he factor of magnification of Q matrix be m, i.e. the new Q matrix is mQ. In Table2 , we vary m from 0.001 to 1000 and illustrate this.... ..."

Cited by 2

### Table 4 Second moments

### Table 5 Second moments

### Table 6 Second moments

### Table 1: Calibration of Health Transition Probability Matrix The flrst column shows the moment, the second column the target from the data, and the last column shows our calibrated value at the chosen parameters. The flrst 8 moments capture aspects related to long-term care (LTC); the data are from Brown and Finkelstein [2004] Table 1 for males. The next 4 moments relate to longevity; the data are from the National Center for Health Statistics, Vital Statistics (1999), Table 2 for males. The last 4 moments show features of the distribution of medical costs. These are not used in the calibration. Details of the calibration exercise are in the appendix. The small discrepancies between the simulation and the data in row 3 arises from the fact that our model is cast in years. The data on the other hand were compiled on a monthly basis. We interpret more than one year as at least two years, and that leads to an upward bias in the average.

2005

"... In PAGE 13: ...TC at $46,700. We take h(3)=46. We ignore costs associated with death by setting h(4) = 0. With these values, the median value for life-time medical expenses is $18K, while the mean is $73K (rows 13 and 14 of Table1 ). Long-term care costs dominate the model, making up 85% of all medical expenses.... In PAGE 41: ...P(0) (of the sixteen elements, four are flxed by the death state being absorbing and there are three further restrictions so that each row sums to one) and three that control the ow of probability from greater health to poorer health as age increases. We calibrate these 12 parameters to match 8 moments related to long-term care utilization (Brown and Finkelstein [2004], Table1 , males), and 4 moments related to longevity (National Center for Health Statistics, Vital Statistics [1999], Table 2 for males). Table 1 shows the moments we match, their target value, and our best flt.... In PAGE 41: ... We calibrate these 12 parameters to match 8 moments related to long-term care utilization (Brown and Finkelstein [2004], Table 1, males), and 4 moments related to longevity (National Center for Health Statistics, Vital Statistics [1999], Table 2 for males). Table1 shows the moments we match, their target value, and our best flt. The last 4 rows show some features of the distribution of medical costs.... ..."

### TABLE 5 PARTIALLY SWEPT MOMENT MATRIX

in KNOWLEDGE REPRESENTATION AND INTEGRATION FOR PORT- FOLIO EVALUATION USING LINEAR BELIEF FUNCTIONS

### TABLE 7 PARTIALLY SWEPT MOMENT MATRIX

in KNOWLEDGE REPRESENTATION AND INTEGRATION FOR PORT- FOLIO EVALUATION USING LINEAR BELIEF FUNCTIONS