Expected utility | Optimal decisions
Generalized expected utility is a decision-making metric based on any of a variety of theories that attempt to resolve some discrepancies between expected utility theory and empirical observations, concerning choice under risky (probabilistic) circumstances. Given its motivations and approach, generalized expected utility theory may properly be regarded as a subfield of behavioral economics, but it is more frequently located within mainstream economic theory. The expected utility model developed by John von Neumann and Oskar Morgenstern dominated decision theory from its formulation in 1944 until the late 1970s, not only as a prescriptive, but also as a descriptive model, despite powerful criticism from Maurice Allais and Daniel Ellsberg who showed that, in certain choice problems, decisions were usually inconsistent with the axioms of expected utility theory. These problems are usually referred to as the Allais paradox and Ellsberg paradox. Beginning in 1979 with the publication of the prospect theory of Daniel Kahneman and Amos Tversky, a range of generalized expected utility models were developed with the aim of resolving the Allais and Ellsberg paradoxes, while maintaining many of the attractive properties of expected utility theory. Important examples were anticipated utility theory, later referred to as rank-dependent utility theory, weighted utility (Chew 1982), and expected uncertain utility theory. A general representation, using the concept of the local utility function was presented by Mark J. Machina. Since then, generalizations of expected utility theory have proliferated, but the probably most frequently used model is nowadays cumulative prospect theory, a rank-dependent development of prospect theory, introduced in 1992 by Daniel Kahneman and Amos Tversky. (Wikipedia).
From playlist Courses and Series
From playlist COMP0168 (2020/21)
The Generalized Likelihood Ratio Test
http://AllSignalProcessing.com for more great signal processing content, including concept/screenshot files, quizzes, MATLAB and data files. There is no universally optimal test strategy for composite hypotheses (unknown parameters in the pdfs). The generalized likelihood ratio test (GLRT
From playlist Estimation and Detection Theory
12 Stochastic Gradient Estimators
Slides and more information: https://mml-book.github.io/slopes-expectations.html
From playlist There and Back Again: A Tale of Slopes and Expectations (NeurIPS-2020 Tutorial)
Statistical Analysis Spreadsheet (1 of 3: Calculating expected values)
More resources available at www.misterwootube.com
From playlist Probability and Discrete Probability Distributions
Expected Value of the Exponential Distribution | Exponential Random Variables, Probability Theory
What is the expected value of the exponential distribution and how do we find it? In today's video we will prove the expected value of the exponential distribution using the probability density function and the definition of the expected value for a continuous random variable. It's gonna b
From playlist Probability Theory
The normal distribution | Probability and Statistics | NJ Wildberger
In this final lecture in this short introduction to Probability and Statistics, we introduce perhaps the most important probability distibution: the normal distribution, also known as the `bell-curve'. Its role is clarified by the Central Limit theorem, a key result in Statistics, that sta
From playlist Probability and Statistics: an introduction
Unit 2 - consumer demand part 4
From playlist Courses and Series
Paolo Guasoni, Lesson II - 19 december 2017
QUANTITATIVE FINANCE SEMINARS @ SNS PROF. PAOLO GUASONI TOPICS IN PORTFOLIO CHOICE
From playlist Quantitative Finance Seminar @ SNS
“Choice Modeling and Assortment Optimization” - Session I - Prof. Huseyin Topaloglu
This module overviews static and dynamic assortment optimization problems. We start with an introduction to discrete choice modeling and discuss estimation issues when fitting a choice model to observed sales histories. Following this introduction, we discuss static and dynamic assortment
From playlist Thematic Program on Stochastic Modeling: A Focus on Pricing & Revenue Management
Eliezer Yudkowsky – AI Alignment: Why It's Hard, and Where to Start
On May 5, 2016, Eliezer Yudkowsky gave a talk at Stanford University for the 26th Annual Symbolic Systems Distinguished Speaker series (https://symsys.stanford.edu/viewing/event/26580). Eliezer is a senior research fellow at the Machine Intelligence Research Institute, a research nonprofi
From playlist Machine Learning Shorts
AI That Doesn't Try Too Hard - Maximizers and Satisficers
Powerful AI systems can be dangerous in part because they pursue their goals as strongly as they can. Perhaps it would be safer to have systems that don't aim for perfection, and stop at 'good enough'. How could we build something like that? Generating Fake YouTube comments with GPT-2: ht
From playlist Technical
Paolo Guasoni, Lesson I - 18 december 2017
QUANTITATIVE FINANCE SEMINARS @ SNS PROF. PAOLO GUASONI TOPICS IN PORTFOLIO CHOICE
From playlist Quantitative Finance Seminar @ SNS
Peter Imkeller: An introduction to BSDE
Abstract: Backward stochastic differential equations have been a very successful and active tool for stochastic finance and insurance for some decades. More generally they serve as a central method in applications of control theory in many areas. We introduce BSDE by looking at a simple ut
From playlist Probability and Statistics
22. Risk Aversion and the Capital Asset Pricing Theorem
Financial Theory (ECON 251) Until now we have ignored risk aversion. The Bernoulli brothers were the first to suggest a tractable way of representing risk aversion. They pointed out that an explanation of the St. Petersburg paradox might be that people care about expected utility inste
From playlist Financial Theory with John Geanakoplos
Multivariate Portfolio Choice via Quantiles
SIAM Activity Group on FME Virtual Talk Series Join us for a series of online talks on topics related to mathematical finance and engineering and running every two weeks until further notice. The series is organized by the SIAM Activity Group on Financial Mathematics and Engineering. Spea
From playlist SIAM Activity Group on FME Virtual Talk Series
Unit 2 - consumer surplus part 2
From playlist Courses and Series
Ellen Vitercik: "How much data is sufficient to learn high-performing algorithms?"
Deep Learning and Combinatorial Optimization 2021 "How much data is sufficient to learn high-performing algorithms?" Ellen Vitercik - Carnegie Mellon University Abstract: Algorithms often have tunable parameters that have a considerable impact on their runtime and solution quality. A gro
From playlist Deep Learning and Combinatorial Optimization 2021