SSRN Author: Robert A. JarrowRobert A. Jarrow SSRN Content
http://www.ssrn.com/author=16130
http://www.ssrn.com/rss/en-usWed, 30 Jul 2014 03:10:02 GMTeditor@ssrn.com (Editor)Wed, 30 Jul 2014 03:10:02 GMTwebmaster@ssrn.com (WebMaster)SSRN RSS Generator 1.0REVISION: Change of Numeraires and Relative Asset Price BubblesIn models of financial bubbles, the price of a stock is a priori typically unbounded, and this plays a fundamental role in the analysis of finite horizon local martingale bubbles. It would seem that price bubbles do not apply to bounded risky asset prices, such as bond prices. To avoid this limitation, to characterize, and to identify bond price mispricings consistent with No Free Lunch with Vanishing Risk, we develop the concept of a relative asset price bubble. This notion uses a risky asset's price as the numéraire instead of the money market account's value. This change of numéraire generates some interesting mathematical complexities because some important numéraires, including risky bonds, can vanish with positive probability over the model's horizon.
http://www.ssrn.com/abstract=2265465
http://www.ssrn.com/1322623.htmlTue, 29 Jul 2014 07:30:16 GMTNew: Optimal Cash Holdings Under Heterogeneous BeliefsThis paper explores a one-period model for a company that finances its operations through debt provided by heterogeneous creditors. Creditors differ in their beliefs about the company's investment outcomes. We analyze the joint equilibrium for the creditors' and the company's optimal investment policies. We show the existence of equilibria in which the company holds cash reserves in order to provide incentives for pessimistic creditors to invest in the company, thereby increasing the debt capacity. This proves the existence of a signaling motive that can help explain the recent high cash holdings puzzle.
http://www.ssrn.com/abstract=2449972
http://www.ssrn.com/1312084.htmlSat, 14 Jun 2014 07:27:26 GMTREVISION: Risk Measures and the Impact of Asset Price BubblesThis paper analyzes the impact of asset price bubbles on a firm's standard risk measures, including value-at-risk (VaR) and conditional value-at-risk (CVaR). Comparing a bubble and non-bubble economy, it is shown that asset price bubbles cause (i) a firm's VaR and CVaR to decline, but (ii) increase its expected daily and maximum daily losses. This decline in the standard risk measures is due to the increased right skew in a firm value's distribution due to bubble expansion. The increase in the expected daily losses is due to bubble bursting. This implies that the standard risk measures are not adequate for equity capital determination in the present of asset price bubbles, and that scenario analysis which include bursting bubbles are essential for the proper determination of equity capital.
http://www.ssrn.com/abstract=2341641
http://www.ssrn.com/1303374.htmlThu, 08 May 2014 03:36:26 GMTNew: Estimating Risk Premia Changes from the Fed's Large-Scale Asset PurchasesThis paper estimates the impact of the Federal Reserve's large-scale asset purchase program on risk premia across the US economy. Whereas much of previous literature focuses upon the direct impact to yields as a result of the Fed's unconventional monetary policies, we study the changes induced to expected returns for both bond and equity investors. Our analysis supports the notion that the Federal Reserve lowered Treasury yields by decreasing the risk premia for interest rate risk. All else equal, this action would contribute to lower expected excess equity returns. However, risk premia for equities actually rose due to the Fed's bond purchases in order to induce investors to shift their holdings from bonds to equities. This rise helps to explain the extraordinary returns realized by equity investors since the financial crisis began in 2008.
http://www.ssrn.com/abstract=2428784
http://www.ssrn.com/1300484.htmlFri, 25 Apr 2014 21:19:25 GMTREVISION: Positive Alphas and a Generalized Multiple-Factor Asset Pricing ModelThis paper derives a generalized multiple-factor asset pricing model using only the assumptions of the existence of an equivalent martingale measure, frictionless, and competitive markets. As such, all existing multiple-factor asset pricing models, including the intertermporal CAPM and Ross' APT, are special cases of this formulation. First, similar to the standard models, a traded asset's expected return is linear in a finite number of traded risk-factor returns. Different from standard models, however, this model allows potentially an infinite number of distinct risk-factors in the economy. Different assets will, in general, depend on a different finite set of risk-factors. Second, positive alphas imply arbitrage opportunities or the existence of dominated securities, and not just abnormal expected returns. This generalization is consistent with many of the observed discrepancies between existing multiple-factor asset pricing models and the empirical evidence.
http://www.ssrn.com/abstract=2368906
http://www.ssrn.com/1298413.htmlThu, 17 Apr 2014 08:28:21 GMTREVISION: Change of Numeraires and Relative Asset Price BubblesIn models of financial bubbles, the price of a stock is a priori typically unbounded, and this plays a fundamental role in the analysis of finite horizon local martingale bubbles. It would seem that price bubbles do not apply to bounded risky asset prices, such as bond prices. To avoid this limitation, to characterize, and to identify bond price mispricings consistent with No Free Lunch with Vanishing Risk, we develop the concept of a relative asset price bubble. This notion uses a risky asset's price as the numéraire instead of the money market account's value. This change of numéraire generates some interesting mathematical complexities because some important numéraires, including risky bonds, can vanish with positive probability over the model's horizon.
http://www.ssrn.com/abstract=2265465
http://www.ssrn.com/1296433.htmlTue, 08 Apr 2014 11:59:46 GMTREVISION: Positive Alphas and a Generalized Multiple-Factor Asset Pricing ModelThis paper derives a generalized multiple-factor asset pricing model using only the assumptions of the existence of an equivalent martingale measure, frictionless, and competitive markets. As such, all existing multiple-factor asset pricing models, including the intertermporal CAPM and Ross' APT, are special cases of this formulation. First, similar to the standard models, a traded asset's expected return is linear in a finite number of traded risk-factor returns. Different from standard models, however, this model allows potentially an infinite number of distinct risk-factors in the economy. Different assets will, in general, depend on a different finite set of risk-factors. Second, positive alphas imply arbitrage opportunities or the existence of dominated securities, and not just abnormal expected returns. This generalization is consistent with many of the observed discrepancies between existing multiple-factor asset pricing models and the empirical evidence.
http://www.ssrn.com/abstract=2368906
http://www.ssrn.com/1284101.htmlSat, 22 Feb 2014 06:35:44 GMTREVISION: Change of Numeraires and Relative Asset Price BubblesIn models of financial bubbles, the price of a stock is a priori typically unbounded, and this plays a fundamental role in the analysis of finite horizon local martingale bubbles. It would seem that price bubbles do not apply to bounded risky asset prices, such as bond prices. To avoid this limitation, to characterize, and to identify bond price mispricings consistent with No Free Lunch with Vanishing Risk, we develop the concept of a relative asset price bubble. This notion uses a risky asset's price as the numéraire instead of the money market account's value. This change of numéraire generates some interesting mathematical complexities because some important numéraires, including risky bonds, can vanish with positive probability over the model's horizon.
http://www.ssrn.com/abstract=2265465
http://www.ssrn.com/1277075.htmlThu, 30 Jan 2014 07:46:25 GMTNew: Computing Present Values: Capital Budgeting Done CorrectlyThis paper shows that the standard textbook formula for computing the present value of a future random cash flow - the discounted expected value - is formally incorrect and can generate significant errors when used to compute present values. The correct present value method is provided as well as a simple adjustment to the textbook formula which can be used to obtain an approximation to the correct value.
http://www.ssrn.com/abstract=2383896
http://www.ssrn.com/1275338.htmlFri, 24 Jan 2014 20:04:33 GMTREVISION: Risk Measures and the Impact of Asset Price BubblesThis paper analyzes the impact of asset price bubbles on a firm's standard risk measures, including value-at-risk (VaR) and conditional value-at-risk (CVaR). Comparing a bubble and non-bubble economy, it is shown that asset price bubbles cause (i) a firm's VaR and CVaR to decline, but (ii) increase its expected daily and maximum daily losses. This decline in the standard risk measures is due to the increased right skew in a firm value's distribution due to bubble expansion. The increase in the expected daily losses is due to bubble bursting. This implies that the standard risk measures are not adequate for equity capital determination in the present of asset price bubbles, and that scenario analysis which include bursting bubbles are essential for the proper determination of equity capital.
http://www.ssrn.com/abstract=2341641
http://www.ssrn.com/1273087.htmlFri, 17 Jan 2014 15:11:17 GMTNew: Informational Efficiency Under Short Sale ConstraintsA constrained informationally efficient market is defined to be one whose price process arises as the outcome of some equilibrium where agents face restrictions on trade. This paper investigates the case of short sale constraints, a setting which despite its simplicity, generates new insights. In particular, it is shown that short sale constrained informationally efficient markets always admit equivalent supermartingale measures and local martingale deflators, but not necessarily local martingale measures. And if in addition some local martingale deflator turns the price process into a true martingale, then the market is informationally efficient. Examples are given to illustrate the subtle phenomena that can arise in the presence of short sale constraints, with particular attention to representative agent equilibria and the different notions of no arbitrage.
http://www.ssrn.com/abstract=2375897
http://www.ssrn.com/1270350.htmlTue, 07 Jan 2014 13:15:50 GMTREVISION: Positive Alphas and a Generalized Multiple-Factor Asset Pricing ModelThis paper derives a generalized multiple-factor asset pricing model using only the assumptions of the existence of an equivalent martingale measure, frictionless, and competitive markets. As such, all existing multiple-factor asset pricing models, including the intertermporal CAPM and Ross' APT, are special cases of this formulation. First, similar to the standard models, a traded asset's expected return is linear in a nite number of traded risk-factor returns. Different from standard models, however, this model allows potentially an innite number of distinct risk-factors in the economy. Different assets will, in general, depend on a different nite set of risk-factors. Second, positive alphas imply arbitrage opportunities or the existence of dominated securities, and not just abnormal expected returns. This generalization is consistent with many of the observed discrepancies between existing multiple-factor asset pricing models and the empirical evidence.
http://www.ssrn.com/abstract=2368906
http://www.ssrn.com/1266685.htmlThu, 19 Dec 2013 03:45:13 GMTNew: Liquidity Suppliers and High Frequency TradingThis paper provides a mathematical analysis of how high frequency traders profit from their speed with respect to the limit order book. We show that their prots can be decomposed into two components. The rest is due to their ability to execute market orders at limit order prices and without incurring any liquidity costs themselves. The second is by "front running" market orders with limit prices. These trading profits are shown to be at the expense of ordinary traders who submit market orders and sophisticated traders who submit limit orders or who use algorithmic trading to split up and execute large trades. We do not consider the welfare implications of our insights to the efficient functioning of financial markets.
http://www.ssrn.com/abstract=2367381
http://www.ssrn.com/1265798.htmlSun, 15 Dec 2013 04:59:23 GMTREVISION: Risk Measures and the Impact of Asset Price BubblesThis paper analyzes the impact of asset price bubbles on a firm's standard risk measures, including value-at-risk (VaR) and conditional value-at-risk (CVaR). Comparing a bubble and non-bubble economy, it is shown that asset price bubbles cause (i) a firm's VaR and CVaR to decline, but (ii) increase its expected daily and maximum daily losses. This decline in the standard risk measures is due to the increased right skew in a firm value's distribution due to bubble expansion. The increase in the expected daily losses is due to bubble bursting. This implies that the standard risk measures are not adequate for equity capital determination in the present of asset price bubbles, and that scenario analysis which include bursting bubbles are essential for the proper determination of equity capital.
http://www.ssrn.com/abstract=2341641
http://www.ssrn.com/1251252.htmlSun, 20 Oct 2013 18:17:36 GMT