Page Not Found
Page not found. Your pixels are in another canvas.
Sitemap
A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.
Pages
Shreeyesh Menon
Shreeyesh Menon
Page not in menu
This is a page not in th emain menu
Teaching
Teaching
Posts
The Geometry of Information
Added:
A geometric interpretation of information
Dynamics: Perspectives from Linear Algebra
Added:
Understanding dynamical systems through linear algebra
Markov Chain Monte Carlo
Added:
Markov Chain Monte Carlo methods are widely used in Bayesian Econometrics. This article aims to clarify their theoretical basis and develop some intuition.
Linear Algebra: Four perspectives
Added:
A quick intro to linear algebra concepts
Singular Value Decomposition
Added:
Singular Value Decomposition and its link with the Pythagoras theorem
Information theory and statistical learning: Some intuition
Added:
Some basics of information theory and entropy. A small discussion on informativeness of signals
Filtering: The Basics
Added:
An introduction to filtering and sensor fusion
Monte Carlo Integration
Added:
Markov Chains: A Gentle Introduction
Added:
Markov Chains are the very foundation of stochastic processes. And they can open up a whole new way of thinking about problems in stochastic processes once we understand them.
portfolio
research
Monetary Policy as Decentralized Control: Near-optimality of Quantized Policy Rules
Added:
This paper studies macroeconomic stabilization as a cooperative game: the central bank and private agents share the common goal of stabilizing the output around potential but act under different information. Policymakers observe private signals about changes in the natural rate of interest that households and firms either lack or find costly to monitor; however, interest rate adjustments are costly making it infeasible to offset them completely. Private agents adjust inflation expectations based on noisy observations on the output gap. Importantly, central bank policy influences private sector’s information by altering aggregate demand and serves as an implicit communication channel, informing them about fundamental shocks allowing for indirect stabilization through adjustment of inflation expectations. This paper draws an analogy of the problem to Witsenhausen’s counterexample, a canonical problem in decentralized control theory, where optimal strategies are inherently non-linear. The analysis shows that quantized policy rules—such as adjusting interest rates in discrete steps—can outperform linear feedback rules like the Taylor rule. Preliminary draft
Download here
Bond market response to FOMC communication: Overreaction to Forward Guidance and Diagnostic Expectations
Added:
Since 2004, the Federal Reserve has released the minutes of Federal Open Market Committee (FOMC) meetings three weeks after each policy announcement, providing a natural setting to study how financial markets process sequential policy signals. This paper examines how the yield curve responds to both FOMC announcements and the subsequent release of meeting minutes. Using high-frequency identification, I extract monetary policy shocks and decompose them into target rate, forward guidance, and yield-curve “twist” components. I document a systematic overreaction of yields to the forward guidance component of announcements, followed by a partial reversal when the corresponding minutes are released. I interpret this pattern through a model of diagnostic expectations, in which investors initially overweight salient forward guidance signals and later revise their beliefs as additional context becomes available in the minutes. Working draft
Keeping up with the Curve: Learning, Evolving Beliefs and the Anchoring of Expectations
Added:
(Job Market Paper) I explain the behavior of inflation, unemployment, and long-run inflation expectations in the post-war U.S. by estimating a forward-looking model in which private agents learn about structural fundamentals and the policymakers’ stabilization preferences in real-time. Learning is structural: agents understand that aggregate economic dynamics result from their optimizing behavior under imperfect knowledge. Monetary policy is conducted optimally but private agents suspect that the policymakers’ stabilization preferences are evolving, which they learn from observed policy behavior. This gives rise to a nonlinear filtering problem. The model provides a novel, information-theoretic explanation of how systematic monetary policy anchors expectations when agents are learning: an increasing emphasis on real-side stabilization has made the Fed’s policy behavior more predictable, stabilizing long-run inflation expectations while at the same time rendering short-run expectations more susceptible to supply shocks. The model offers a new explanation for the recent post-Covid inflation surge and the ``costless disinflation’’ that followed, shedding light on why it differed markedly from the crisis of the 1970s and 1980s. Download paper
Download here