Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore the mathematical analysis of first crossing times in auto-regressive Markov chains through this 53-minute research lecture from Institut des Hautes Etudes Scientifiques. Examine the investigation of zero-crossing behavior in auto-regressive Markov chains with atomless innovations, focusing on the random variable T representing the first crossing time. Learn about the log-concavity assumptions applied to innovation laws and discover how these conditions lead to log-convex probability distributions for positive drifts, establishing connections to Baxter-Spitzer factorization similar to random walk theory. Understand the contrasting behavior observed with negative drifts, where log-convexity properties break down completely. Delve into advanced conjectures regarding complete monotonicity of probability laws for positive drifts and the potential relationship between discrete Baxter-Spitzer factorization and continuous Wiener-Hopf factorization. Gain insights into cutting-edge research in probability theory and stochastic processes from Thomas Simon of Université de Lille, presented as part of the mathematical research community's ongoing exploration of Markov chain persistence phenomena.
