643 reputation
18
bio website ventirisk.com
location State College, PA
age 49
visits member for 1 year, 9 months
seen Jun 5 at 21:29

Arthur Small is co-founder and CEO of Venti Risk Management, a consultancy and software development firm based in State College, Pennsylvania, USA. His work focuses on the management of risks arising due to variations in weather and climate.

Before co-founding Venti, Small served as Associate Professor of Applied Economics and Finance in the Department of Meteorology at Penn State University; and as Assistant Professor in the Graduate School of Business and the School of International and Public Affairs at Columbia University. He holds a A.B. degree in mathematics from Columbia University, an M.S. in mathematics from Cornell University, and a Ph.D. in Agricultural and Resource Economics from the University of California at Berkeley.


Nov
28
awarded  Yearling
Mar
24
comment Is a subset $B= {(-\infty,t),t\in \mathbb{R} }$ an event of sample space $\mathbb{R}$?
@Zen is correct, and I was mistaken. I have edited my answer accordingly.
Mar
24
revised Is a subset $B= {(-\infty,t),t\in \mathbb{R} }$ an event of sample space $\mathbb{R}$?
Amended answer to correct an error identified by another user; fixed grammar.
Mar
23
answered Is a subset $B= {(-\infty,t),t\in \mathbb{R} }$ an event of sample space $\mathbb{R}$?
Feb
3
answered Creating conditional probability distribution
Jan
30
comment Determine likelihood that a list of N items contains no negative results after examining X items
For an answer to a question that is formally similar to yours, see: stats.stackexchange.com/questions/47286/…
Jan
20
comment Measuring representativeness of a (non-random) selection
@Thilo: Oh, dear. I suppose I invited that follow-up, so have only myself to blame. Answering would, alas, take a long while. 'Any' stats textbook will cover the topic: see, e.g., W.H. Greene, Econometric Analysis. On 'why not?', I'll recommend a book I find canonical that covers the philosophical and practical problems of hypothesis testing quite well and in depth: Statistical Decision Theory and Bayesian Analysis by J.O. Berger, amazon.com/gp/product/1441930744. Very short answer: I'm interested in using stats to make better decisions under uncertainty. HT doesn't help.
Jan
17
comment Measuring representativeness of a (non-random) selection
There is a noteworthy economic literature on statistical measurement of race and sex discrimination in, e.g., labor markets, that might be useful. A seminal work is: The Statistical Theory of Racism and Sexism Edmund S. Phelps The American Economic Review Vol. 62, No. 4 (Sep., 1972), pp. 659-661 Published by: American Economic Association Article Stable URL: jstor.org/stable/1806107
Jan
17
revised Measuring representativeness of a (non-random) selection
In example: Replaced hand-waving approximate odds ratio with exact calculation; added brief proviso about election procedures for U.S. senators.
Jan
17
revised Measuring representativeness of a (non-random) selection
Added interpretation of the numerical result as an odds ratio.
Jan
17
answered Measuring representativeness of a (non-random) selection
Jan
16
comment Minimizing variance of an estimator under sampling cost penalty
@StasK: Touche$'$ :)
Jan
16
awarded  Tag Editor
Jan
16
revised value-of-information wiki excerpt
added 138 characters in body
Jan
16
wiki created value-of-information excerpt
Jan
16
suggested suggested edit on value-of-information tag wiki excerpt
Jan
16
comment Time-series stationarity
Differencing is most useful for analyzing series with unit roots, e.g., AR processes with coefficients equal to (or close to) unity. Consider the simplest such case in which $y_t$ follows a random walk, i.e., in which $y_t = y_{t-1} + e_t$. This process is non-stationary. However, the differenced series is stationary. Ref: en.wikipedia.org/wiki/Unit_root
Jan
16
revised Minimizing variance of an estimator under sampling cost penalty
Fixed statement of an equation.
Jan
16
comment Minimizing variance of an estimator under sampling cost penalty
@Jugurtha: My bad: I should have written $E[C \mid \cdot]$ as a double integral: $E[C \mid \cdot] = \int \int C(\hat{\theta},\theta) f(\hat{\theta} \mid x_{1:n}) d\hat{\theta} f({\theta} \mid x_{1:n}) d\theta$. That is, given a beliefs about the distribution of $\theta$, you are able to formulate expectations of how far off a given point estimate is likely to be. (I'll edit my original answer to fix.)
Jan
16
comment R vs SAS, why is SAS preferred by private companies?
@DimitriyV.Masterov: Revolution R is a commercial "enterprise" version of R that claims to allow R to run in-database. (I have not reviewed or audited this claim.) Revo R is not free ($1,000, I think), but its much cheaper than SAS.