"... In PAGE 11: ... Following the discussion in Sectio n 3.3, Table4 shows the main sources of common IT risks mapped to specific operating options that can control them (e.g.... In PAGE 13: ...13 The recognized shadow options must be examined carefully. Table4 might map a particular risk to certain options that can control it, but not every investment exposed to that risk would necessarily embed all those options. Some recognized op tions would be immediately viable if they are already embedded in the target investment (e.... In PAGE 13: ... In other words, not every recognized shadow option is necessarily viable. For example, Table4 could suggest the abandon option as a way to control market risk, but regulatory constraints could prohibit killing the target investment shortly aft er it becomes operational. Table 5 lists key conditions that must be met in order for operating options of different types to be viable.... In PAGE 13: ...ment, invest partially, etc.) and in the investment context. Note that these conditions exclude option - cost and option - value considerations; these considerations are treated in step 4. ISC Illustrat ion (continued) The gray cells in Table4 show the shadow options recognized in the ISC investment. The options are: defer, stage, explore, outsource, contract, expand, and abandon.... In PAGE 15: ... Conceptually, steps 1 through 3 construct a decision tree that encompasses every plausible form of managerial flexibility. Starting with a plain decision tree reflecting the NPV P , and given the risks present at each stage in the investment lifecycle, we track through the tree and use Table4 to suggest which decision nodes (or options) can (re)configure the investment in w ays that suppress and/or enhance poor and/or good outcomes. Each path through this tree includes a different subset of these decision nodes (options), corresponding to one viable configuration.... ..."

"... In PAGE 7: ... Each cohort which embarks on equity investment at age 25 faces a di erent 35 years of returns on the equity market. In Table4 , we focus on the 25th and 75th percentile of the distribution of the terminal wealth. We see a fairly large variability associated with Strategy B.... In PAGE 7: ... Table 3 shows that there is a fairly small risk that the all-equity strategy could underperform the all-bonds strategy. However, Table4 re- veals a wide spread in the range of outcomes that could derive from the all{equity strategy. What can be done in containing this investment risk, and in obtaining a more predictable outcome? 6.... In PAGE 8: ... The results from this strategy are summarised in Table 5. A comparison with Table4 reveals that this strategy routinely obtains a much lower inter{ quartile range; however a comparison with Table 2 reveals signi cantly lower accumulation. As emphasised by Alier amp; Vittas (1999), this strategy { of phasing out equity exposure { has the appeal of being easily implemented using unso- phisticated nancial markets.... ..."

"... In PAGE 5: ...evelopment is completed and the application becomes operational (e.g., regulatory actions). Table1 offe rs a high - level synthesis of the central forms of IT investment risks identified by both research streams. As suggested earlier, these forms of risk can be placed into three categories.... In PAGE 6: ...Table1... ..."

"... In PAGE 7: ... Table1 : The complementary operations of classification and construction and their respective properties. 4 Ecosystems Various extreme situations from the evolution of life will be used as paradigmatic examples for which all the abstract notions introduced above can be applied.... ..."

"... In PAGE 5: ... The learner model is represented in the form of classification rules, decision trees or mathematical formulae. Table4 illustrates a step by step approach of classification: Decision trees can be used to classify new cases. They can construct explicit symbolic rules that generalize the training cases (rule induction and decision tree induction).... ..."