Kenneth Roskelley

Asymmetric Revisions to Primary and Secondary Shares in Seasoned Equity Offerings

This paper examines the relation between revealed market demand and insider amendments to the number of primary and secondary shares offered during the seasoned equity offering (SEO) process. We define strong (weak) market demand for an SEO as a positive (negative) buy-and-hold abnormal return over the registration window. We find that insiders make amendments to primary and secondary shares conditional on the market demand revealed in the SEO registration period, but that the frequency, direction, and size of the amendments are not consistent across the two types of shares. For instance, insiders amend the prospectus to significantly decrease (increase) the number of secondary shares when the revealed market demand during the registration period is weak (strong). Amendments to originally filed primary shares are less prevalent. In addition, downward revisions in primary shares are unrelated to the market demand, and are less likely to occur the higher the pre-announcement stock price run-ups. Upward revisions in primary shares, however, are sensitive to strong market demand revealed in the registration period. Finally, when insiders amend the prospectus to increase (decrease) secondary shares, they either keep the number of primary shares constant, or increase (decrease) secondary shares disproportional to primary shares. These results are consistent with insiders employing a demand-conditioned adjustment strategy to secondary shares, as well as acting opportunistically to maximize personal wealth.

Cromwell's Rule and the Role of the Prior in the Economic Metric

We show that using Bayesian decision theory to draw inference about the economic significance of a model requires careful specification of model uncertainty in the prior density. As an example, we use the Bayesian investors portfolio allocation problem to show that failure to include probability point mass on the null hypothesis that returns are not predictable will overstate the economic significance of the predictive model. The dissonance between the statistical and economic significance of asset return predictability seen in previous research appears to be an artifact of how model uncertainty is treated in the specification of the prior.

Testing Arbitrage Models Without Adding an Error Model: An Application to Cox, Ingersoll, and Ross

We demonstrate a simple procedure to test arbitrage models without adding an auxiliary error model. Our tests rely on the dynamics of the model to draw inference through out-of-sample forecasting. As an illustration, we estimate the Cox et al. model with a rolling sample to forecast zero-coupon yields from 1994 to 2007. We use these forecasts to test the model as both a candidate return generating process and to assess its efficacy as part of a forecasting method. The model is soundly rejected. Since our empirical design maintains the models stochastic singularity, the affine term structure models poor empirical performance cannot be blamed on an unfortunate choice of an auxiliary error model. Unlike earlier studies, the traditional expectations hypothesis holds in our sample, and the model cannot reproduce this feature of the data.

Empirically Confronting Stochastic Singularity: An Application to the Cox, Ingersoll, and Ross Model

We just-identify a no-arbitrage term structure model in estimation and then test it using both a classical orthogonality restriction test and a test of conditional predictive ability. We treat the error structure as unmodeled heterogeneity so that the model is estimated without errors, and the statistical question is whether using the model to characterize the dynamics and patterns in historical data is either useful or optimal as a forecasting tool. The data we use are from the transparent Greenspan regime at the Fed (1989-2005), and we also use a rolling estimation format so that regime shifts are not a likely cause of the models performance. Substantively we find that the model is not a good forecasting device for short rates which in this period are strongly affected by changes in the target Fed Funds rate. For longer term rates, especially at longer forecast horizons where Fed policy has no effect, the model is more informative.

Cromwell's Rule and the Role of the Prior in the Economic Metric: An Application to the Portfolio Allocation Problem

I show that using Bayesian decision theory to draw inference about the economic significance of a model requires careful specification of model uncertainty in the prior density. As an example, I use the Bayesian investors portfolio allocation problem to show that failure to include probability point mass on the null hypothesis that returns are not predictable will overstate the economic significance of the predictive model. The dissonance seen in previous research between the statistical and economic significance of asset return predictability appears to be an artifact of how model uncertainty is treated in the specification of the prior.