22.12.2012 Views

of 3 - Center for Global Outsourcings

of 3 - Center for Global Outsourcings

of 3 - Center for Global Outsourcings

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

If n increases, the <strong>for</strong>ecasting errors converge, but also increase costs <strong>of</strong> hiring additional<br />

managers. The optimal n depends on the firm and/or industry. If q percent are optimistic and (1q)<br />

are pessimistic, the joint <strong>for</strong>ecasted probability is: ( ) (∑<br />

∑<br />

( )<br />

). For a given q, error converges to a stable point with increase in n. Also, higher q<br />

results in a smaller error (i.e. | | , be<strong>for</strong>e q=1) (Figure 2a).<br />

Figure 2a Forecasting Errors Given q Known (left) and Figure 2b (q random), right figure<br />

If q is randomly generated, the net <strong>for</strong>ecasting errors diverge (Figure 2b).<br />

Summary and Conclusions<br />

Firms can reduce managerial anchoring costs if a committee approach is adopted (if managers<br />

anchor in different directions), and/or when future cash flows are volatile. However, higher<br />

marginal costs <strong>of</strong> hiring can <strong>of</strong>fset lower anchoring costs. Future research can remove model<br />

assumptions and can consider sequential ‘learning’ by manager committee members.<br />

Tenth Annual International Daejeon, South Korea P a g e | 42<br />

Smart Sourcing Conference June 28-29, 2012

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!