Logistic regression
Housing statistics and English Housing Survey, glossary, published by the Department for Levelling Up, Housing and Communities in 2019, defines logistic regression as: ‘…a regression model where the dependent variable is binary i.e. takes one of two values which are assigned as 0 or 1. The model predicts the probability of the dependent variable taking the value 1 for particular values of the independent variables. The regression coefficients are usually estimated using maximum likelihood. Logistic regression measures the relationship between the categorical dependent variable and one or more independent variables by estimating probabilities using a logistic function, which is the cumulative logistic distribution. Thus, it treats the same set of problems as probit regression using similar techniques, with the latter using a cumulative normal distribution curve instead. Equivalently, in the latent variable interpretations of these 2 methods, the first assumes a standard logistic distribution of errors and the second a standard normal distribution of errors.’
[edit] Related articles on Designing Buildings
Featured articles
Check out some of the best features and news from Designing Buildings as well as key stories from around the web.
ECA responds to the Milburn Report
Youth employment to be linked to workforce demand.
The construction industry urgently needs to attract new talent
CIOB responds to Neets report.
Grenfell Tower community update
Memorial columns to be retained as careful dismantling continues.
When buildings reflected the local resources available.
CIOB reports 57% are using artificial intelligence technologies.
More of England’s ‘hidden heroes’ will be celebrated.
CIAT calls for retrofit to be a priority.
Version 6 is now available, along with guidance for ECA members.
Post-War Homes: mid-century British architecture. Book review.
CIAT reflects on the party conferences.
CIOB announce the winners of the Art of Building photography competition.
National Scheme of Delegation introduced.

















Comments
[edit] To make a comment about this article, or to suggest changes, click 'Add a comment' above. Separate your comments from any existing comments by inserting a horizontal line.