Search
× Search
Saturday, December 21, 2024

Archived Discussions

Recent member discussions

The Algorithmic Traders' Association prides itself on providing a forum for the publication and dissemination of its members' white papers, research, reflections, works in progress, and other contributions. Please Note that archive searches and some of our members' publications are reserved for members only, so please log in or sign up to gain the most from our members' contributions.

A Winer Process

photo

 Jonathan Kinlay, Quantitative Research and Trading | Hedge Fund Partner & Leading Expert in Quantitative Algorithmic Trading Strategies

 Monday, September 25, 2017

A look at some of the machine learning capabilities in the latest version of Wolfram Mathematica. https://www.linkedin.com/pulse/winer-process-jonathan-kinlay?published=t


Print

6 comments on article "A Winer Process"

photo

 Vasily Nekrasov, Senior Risk Analyst and Model Developer at Total Energie Gas GmbH

 Sunday, October 1, 2017



But what is wine quality indeed? :)

How to measure it quantitatively?


photo

 Jonathan Kinlay, Quantitative Research and Trading | Hedge Fund Partner & Leading Expert in Quantitative Algorithmic Trading Strategies

 Sunday, October 1, 2017



There are several well established metrics. For example: https://www.robertparker.com/ratings


photo

 Raphael Douady, Professor of Applied Mathematics at Stony Brook University

 Monday, October 2, 2017



Just as love, passion, intelligence or happiness, plenty of metrics available!

My friend, the day you can measure the quality of a wine and eliminate subjectivity, thar day, humanity will really have disappeared ! Even robots are subjective. Even if people are completely replaced by robots and disappear, robots will disagree on wine quality...


photo

 Jonathan Kinlay, Quantitative Research and Trading | Hedge Fund Partner & Leading Expert in Quantitative Algorithmic Trading Strategies

 Tuesday, October 3, 2017



Well, here's the thing: it may not matter. A fully objective measure of quality may not be necessary, as long as you don't care whether the value system is shared by others.

Presumably you know your own taste in wine and, if you drink it consistently, can fairly reliably determine which wines you like or don't like. From there you can develop your own scoring system. If it correlates well with one of the existing wine ranking measures, you are good to go. If not, you will need to train a network to extract features that allow you to classify wines according to your own, specific taste.

I don't think its that hard.


photo

 Dimitry Murzinov, Quantitative Risk Management

 Wednesday, October 4, 2017



Both low acidity ( high ph) and high alcohol are due to perfectly ripe grapes. And this is where climate comes in . Look for warm region with hot nighttime temperatures and longer growing season. No Riesling then any more 😁


photo

 MARTIN SMIETANSKI, ceo at GLOTEX

 Wednesday, October 4, 2017



My proposition is consumer price metrics based on ratio/distribution between price (rated from 1-100) and number of bottles sold. you can improve this with zip-code classifier as people in 90210, 94558 probably know more about wine then in 11222. Credit card companies processing could have good transaction record of actual think - this however suffer errors derived by over-marketing/ regional economic dependency and shortages in supply - so that would be the start as in the end I imagine you still want sales pricing / forecasting model...

Please login or register to post comments.

TRADING FUTURES AND OPTIONS INVOLVES SUBSTANTIAL RISK OF LOSS AND IS NOT SUITABLE FOR ALL INVESTORS
Terms Of UsePrivacy StatementCopyright 2018 Algorithmic Traders Association