Every number on this page is computed by the Chapter 3 notebook's own code. The book declares no peer companies, so the two used here are drawn with the lab's random seed 26103 and labelled as such. Your notes stay on this page until you leave it; copy the memo before you go.
Meridian's board has asked two questions that sound simple: which of two peer companies is Meridian most like, and does it matter in which order the restructuring and the digital program are carried out? You are on the corporate development team, and you suspect that the answers depend on choices—of units, of distance measure, of sequence—that nobody has made explicitly.
Use AXIOM to see how units and distance measures shape comparisons between companies, and how the order of two transformations shapes the company that results.
Meridian's eight readings, each with its unit (Table 3.1): the canonical starting state for both volumes.
In the vector bench, read Meridian's eight readings and their units. Switch the liquidity unit from billions to millions, then record net leverage as a percentage rather than a multiple, and watch the raw distances to the two peer companies. Does the closer peer change?
Switch on the declared scaling and compare again. Which comparison would you defend to the board, and why?
The declared scaling divides each unbounded reading by today's value and each index by its range of 100 (Section 3.3; Figure 3.3). Bars: each reading's share of the squared distance to each peer.
In the norm gallery, compare the effort, bottleneck, and Euclidean views of the same change of state, and note the constants that link them. When would a board care about the largest single move rather than the total movement?
Changes of state in the declared scaling: each option's move to its Year-5 state in the Chapter 1 model (Table 1.9), and each peer. Effort ‖·‖₁ totals the moves, bottleneck ‖·‖∞ is the largest single move, Euclidean ‖·‖₂ sits between (Definition 3.5). For eight readings, ‖x‖∞ ≤ ‖x‖₁ ≤ 8‖x‖∞ and ‖x‖₂ ≤ ‖x‖₁ ≤ √8 ‖x‖₂ (Theorem 3.1; Online Appendix 3.B).
In the operator composer, apply the restructuring and the digital investment in both orders. Record the two end states and the size of the difference between them.
The sale brings in $112 million net, takes 10 percent of the technology position and, unless the contract keeps the shared engineering team, 5 points of workforce capability. The digital program is Worked Example 1.1's build: cash first falls to $138 million below its start around month 21 and ends $35 million down, while the platform rises 34.6 points; it needs workforce capability of at least 55 (Worked Example 3.1).
Size of the difference between the two feasible orders, in the declared scaling (liquidity over $620 million, indices over 100).
On the linear dynamics stage, run the three-reading company of Section 3.4 with no investment, then with investment, and read the eigenvalue display.
Equation 3.2: xk+1 = A xk + B uk for workforce capability, technology platform and operational efficiency, from (50, 40, 60). Section 3.4 runs one year at an investment of 10; Chapter 7 holds 8 a year.
A change of units can reorder raw comparisons until the scaling is declared (Proposition 3.2); the distance measures disagree on numbers but not on conclusions (Theorem 3.1); and the two orders end in different places (Theorem 3.2).
Recommend the order of the two initiatives, with the end states that support it and one sentence on the units and scaling used.