Forecasting: the essential skills is aimed at improving forecasting skills.  Chapters include the costs of forecasting errors; seven forecasting case studies; an evaluation of the performance of economic forecasting; an evaluation of the performance of exchange rate forecasting; an overview of forecasting techniques; risk and ncertainty in forecasting; an extended description of forecasting with the wisdom of the masses (including a case study) and the essential forecasting skills.  The author, Charles J Nelson is a statistician who has over 40 years of forecasting experience spanning a range of fields.

Display per page
Sort by

This book aims to improve the accuracy of forecasting. The book is about forecasting skills in two senses: Forecasting skill is a measure of the extent to which forecasts are more accurate than those produced by naive methods such as trend extrapolation. The book aims to improve skill in this sense. There is a range of skills needed by the individual or team making forecasts and the book identifies these skills through evaluation of forecasting performance in several fields and through case studies. The skills span a wide range including knowledge, experience, and ways of thinking.

Published in May 2018, the book has 19 chapters, 24 case studies, over 170 pages and 50,000 words.

Now comes with Volume 2, Part 1 written in 2020, which analyses perceptions of forecasting accuracy.


There are periods when GDP forecasts are most likely to be inaccurate.  This report identifies such times and the factor which causes inaccuracy.


What does 2021 hold for Australia?  We have consulted the Wisdom of the Masses to find out.


Opinion polls conducted before the 2019 federal election all got it wrong by a similarly large margin.  An investigation by #AMSRO (Association of Market and Social Research Organisations) concluded that:

“The most likely reason why the polls underestimated the first preference vote for the LNP and overestimated it for Labor was because the samples were unrepresentative and inadequately adjusted.

The polls were likely to have been skewed towards the more politically engaged and better educated voters with this bias not corrected.

As a result, the polls over-represented Labor voters.

Pollsters share a common difficulty in struggling to establish contact with and gain the cooperation of a representative sample of voters – irrespective of methodology.  This conclusion is broadly similar to that reached by the reviews into the performance of the 2015 UK polls and the 2016 US polls”.


After this excoriating assessment, how can we more accurately predict the outcome of the next federal election, which will most likely be held between August 2021 and April 2022?