Investment-Cash Flow Sensitivity and Proxy Quality Thresholds

Toni M. Whited, Timothy Erickson
2003 Social Science Research Network  
We revisit the problem of using error-laden proxies for q in regressions of investment on cash flow and q. We give a menu of prior information sets that can identify the sign of the coefficient on cash flow. This information is weaker than that necessary to identify coefficient values. An element of each set is a prior lower bound on a correlation between the proxy and q. We obtain positive OLS estimates of the coefficient on cash flow. However, the partial correlation between observed and true
more » ... n observed and true q must be unrealistically large to infer a positive cash-flow coefficient in the absence of measurement error. J.E.L. classification: G31; E22; C29; C49
doi:10.2139/ssrn.354700 fatcat:k6looaqo3becdkvx3shejgpgfu