"... In PAGE 2: ... Assuming that Table 2 below is correct, the risk type for four of 69 HPVs is not known, so that 65 HPVs are evaluated. Twenty among 65 HPVs are classified as high-risk and the remaining 45 are classified as low-risk, while there are only 12 high-risk HPVs in Table2 . Since 53 HPVs are correctly classified, the accuracy is 81.... In PAGE 4: ... Lastly, we used the comment of HPV types to classify some types difficult to be classified. Table2 shows the summarized classification of HPVs according to its risk. In the all experiments below, we used only lt;comment gt; part.... In PAGE 6: ... Table 5 shows the predicted risk type for the HPV types whose risks are not known exactly. These HPVs are described as null ?null in Table2 . According to previous research on HPV (Chan et al.... ..."

"... In PAGE 21: ...5 The Risk Characteristics of Pairs Trading Strategies. To provide further perspective on the risk of pairs trading, Table4 compares the risk premium of pairs trading to the market premium (SP500), and reports the risk-adjusted returns to pairs trading using two different models for measuring risk. Table 5 summarizes value-at-risk measures for pairs portfolios.... In PAGE 21: ... Table 5 summarizes value-at-risk measures for pairs portfolios. The top part of Table4 compares the excess return to pairs trading to the excess return on the SP500. Between 1963 and 2002, the average excess return to pairs trading has been about twice as large as the excess return of the SP500, with only one half to one third of the risk as ... In PAGE 22: ... To examine this possibility, we include a momentum factor in our risk model that is constructed along the lines of Carhart (1997). Table4 shows that only a small portion of the excess returns of pairs trading can be attributed to their exposures to the five risk factors. The intercepts of the regressions show that risk-adjusted returns are significantly positive, and lower than the raw excess returns by about 10-20 bp per month.... In PAGE 23: ... The next sections provide further evidence to support this view. The bottom of Table4 shows the results of regressing the pairs returns on an alternative set of risk factors suggested by Ibbotson: the excess return on the S amp;P 500, the U.S.... In PAGE 25: ...6 Pairs trading and contrarian investment Because pairs trading bets on price reversals, it is an example of a contrarian investment strategy. The results of Table4... ..."

"... In PAGE 7: ... Think of S as investing in baggage screening, and N as not doing so. Table1 shows the payoffs to the agents for the four possible outcomes: Table 1: Expected Costs Associated with Investing and Not Investing in Security ... In PAGE 8: ... So the risk of contagion only matters to an agent in the case in which that agent does not suffer damage originating at home. Now that the payoffs have been specified, we can ask the natural question: under what conditions will the agents invest in security? It is clear from Table1 that for investment in security to be a dominant strategy, we need Y-c gt;Y-pL and Y-c-qL gt;Y-pL-( 1-p) qL The first inequality just says that c lt;pL, that is, the cost of investing in security must be less than the expected loss, a natural condition for an isolated agent. The second inequality is more interesting: it reduces to c lt;pL- pqL = pL( 1-q) .... In PAGE 9: ...2 , L=1000 and c= 95. The matrix in Table1 is now represented as Table 2. Table 2: Expected Costs Associated with Investing and Not Investing in Security for Illustrative Example Agent 2 ( A2 ) S N S Y-95, Y-95 Y-295, Y -100 Agent 1 ( A1 ) N Y-100, Y-295 Y-280, Y- 280 One can see that if A2 has protection (S), then it is worthwhile for A1 to also invest in security since its expected losses will be reduced by pL= 100 and it will only have to spend 95 on the security measure.... In PAGE 12: ... So from Figure 1 we see that the upper bound on c rises with n and the incentive to invest increases. What is the structure of the set of possible Nash equilibria? Return to Table1 . As we have already seen, (S,S) is a dominant strategy equilibrium if c lt;pL(1-q).... ..."

"... In PAGE 5: ... Think of S as investing in baggage screening, and N as not doing so. Table1 shows the payoffs to the agents in the four possible outcomes: Table 1: Expected Costs Associated with Investing and Not Investing in Security ... In PAGE 6: ... So the risk of contagion only matters to an agent in the case in which that agent does not suffer damage originating at home. Now that the payoffs have been specified, we can ask the natural question: under what conditions will the agents invest in security? It is clear from Table1 that for investment in security to be a dominant strategy, we need Y-c gt;Y-pL and Y-c-qL gt;Y-pL-( 1-p) qL The first inequality just says that c lt;pL, that is, the cost of investing in security must be less than the expected loss, a natural condition for an isolated agent. The second inequality is more interesting: it reduces to c lt;pL- pqL = pL( 1-q) .... In PAGE 6: ...2 , L=1000 and c= 95. The matrix in Table1 is now represented as Table 2. ... In PAGE 9: ... Next we discuss the structure of the set of possible Nash equilibria. Return to Table1 . As we have already seen, (S,S) is a dominant strategy equilibrium if c lt;pL(1-q).... ..."

"... In PAGE 12: ...Table2... ..."